customModel_DenseNet121_BS_16_25Nov2021.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# COVID-19, normal (healthy), and pneumonia classification using Deep Learning Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DenseNet121 with image size 256x256, categorical cross-entropy loss function and 16 batch size, trained over 1000 epochs to obtain the best model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import the necessary libraries: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25_Nov_2021--03_06_10_AM\n"
     ]
    }
   ],
   "source": [
    "from datetime import datetime\n",
    "dt=datetime.now().strftime(\"%d_%b_%Y--%I_%M_%S_%p\")\n",
    "print(dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import the necessary packages\n",
    "import os\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import regularizers\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras import Model\n",
    "from tensorflow import keras\n",
    "import matplotlib.pyplot as plt\n",
    "#from tensorflow import keras\n",
    "from tensorflow.keras.layers import AveragePooling2D, Dropout, Flatten, Dense, Input\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from tensorflow.keras import optimizers, models, layers\n",
    "from tensorflow.keras.preprocessing import image\n",
    "from tensorflow.keras.applications import VGG16\n",
    "from tensorflow.keras.applications.inception_v3 import InceptionV3\n",
    "from tensorflow.keras.applications.resnet50 import ResNet50\n",
    "from tensorflow.keras.applications.densenet import DenseNet121\n",
    "\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "from imutils import paths\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import cv2\n",
    "import os\n",
    "from tensorflow.keras.models import Sequential,Model\n",
    "from tensorflow.keras.layers import Conv2D, Flatten, MaxPooling2D, Dense, Input, Reshape, Concatenate, GlobalAveragePooling2D, BatchNormalization, Dropout, Activation, GlobalMaxPooling2D\n",
    "from tensorflow.keras.utils import Sequence\n",
    "from tensorflow.keras.applications.densenet import DenseNet121\n",
    "#from keras.callbacks import EarlyStopping\n",
    "from tensorflow.keras.utils import plot_model\n",
    "#from keras.utils.vis_utils import pydot\n",
    "import re\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "import seaborn as sn\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_dir = ''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "''"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base_dir"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dir = os.path.join(base_dir, 'newtrain') # change here newtrain to train\n",
    "validation_dir = os.path.join(base_dir, 'validation')\n",
    "\n",
    "train_COVID_dir = os.path.join(train_dir, 'covid')\n",
    "train_NORMAL_dir = os.path.join(train_dir, 'normal')\n",
    "train_PNEUMONIA_dir = os.path.join(train_dir, 'pneumonia')\n",
    "\n",
    "validation_COVID_dir = os.path.join(validation_dir, 'covid')\n",
    "validation_NORMAL_dir = os.path.join(validation_dir, 'normal')\n",
    "validation_PNEUMONIA_dir = os.path.join(validation_dir, 'pneumonia')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "newtrain\\covid\n",
      "newtrain\\normal\n",
      "newtrain\\pneumonia\n",
      "\n",
      "validation\\covid\n",
      "validation\\normal\n",
      "validation\\pneumonia\n"
     ]
    }
   ],
   "source": [
    "print(train_COVID_dir)\n",
    "print(train_NORMAL_dir)\n",
    "print(train_PNEUMONIA_dir)\n",
    "print()\n",
    "print(validation_COVID_dir)\n",
    "print(validation_NORMAL_dir)\n",
    "print(validation_PNEUMONIA_dir)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['cov_1110.jpg', 'cov_1111.jpg', 'cov_11110.jpg', 'cov_11111.jpg', 'cov_11112.jpg', 'cov_11113.jpg', 'cov_11114.jpg', 'cov_11115.jpg', 'cov_11116.jpg', 'cov_11117.jpg']\n",
      "\n",
      "['normal-0000.jpg', 'normal-0001.jpg', 'normal-00010.jpg', 'normal-00011.jpg', 'normal-00012.jpg', 'normal-00013.jpg', 'normal-00014.jpg', 'normal-00015.jpg', 'normal-00016.jpg', 'normal-00017.jpg']\n",
      "\n",
      "['pneumonia0000.jpg', 'pneumonia0001.jpg', 'pneumonia00010.jpg', 'pneumonia00011.jpg', 'pneumonia00012.jpg', 'pneumonia00013.jpg', 'pneumonia00014.jpg', 'pneumonia00015.jpg', 'pneumonia00016.jpg', 'pneumonia00017.jpg']\n",
      "total training COVID images : 33\n",
      "total training NORMAL images : 37\n",
      "total training PNEUMONIA images : 22\n",
      "total validation COVID images : 1432\n",
      "total validation NORMAL images : 1000\n",
      "total validation PNEUMONIA images : 1000\n"
     ]
    }
   ],
   "source": [
    "train_COVID_fnames = os.listdir( train_COVID_dir )\n",
    "train_NORMAL_fnames = os.listdir( train_NORMAL_dir )\n",
    "train_PNEUMONIA_fnames = os.listdir( train_PNEUMONIA_dir )\n",
    "\n",
    "print(train_COVID_fnames[:10])\n",
    "print()\n",
    "print(train_NORMAL_fnames[:10])\n",
    "print()\n",
    "print(train_PNEUMONIA_fnames[:10])\n",
    "\n",
    "print('total training COVID images :', len(os.listdir(      train_COVID_dir ) ))\n",
    "print('total training NORMAL images :', len(os.listdir(      train_NORMAL_dir ) ))\n",
    "print('total training PNEUMONIA images :', len(os.listdir(      train_PNEUMONIA_dir ) ))\n",
    "\n",
    "print('total validation COVID images :', len(os.listdir( validation_COVID_dir ) ))\n",
    "print('total validation NORMAL images :', len(os.listdir( validation_NORMAL_dir ) ))\n",
    "print('total validation PNEUMONIA images :', len(os.listdir( validation_PNEUMONIA_dir ) ))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Experimental parameters "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set learning rate, epochs and batch size\n",
    "INIT_LR = 1e-5    # This value is specific to what model is chosen: Inception, VGG or ResNet etc.\n",
    "EPOCHS = 2 # changed to 1000\n",
    "BS = 16\n",
    "imagesize=256"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the sample image files provided here "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "give the path totraining dataset in this variable\n",
    "imagePath = \"./train/\"  # change to your local path for the sample images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading images...\n",
      "... ...  92 images loaded in multiple classes:\n",
      "['./newtrain/covid' './newtrain/normal' './newtrain/pneumonia']\n"
     ]
    }
   ],
   "source": [
    "print(\"Loading images...\")\n",
    "imagePath = \"./newtrain/\"  # change to your local path for the sample images\n",
    "#imagePath = \"./train/\"  # change to your local path for the sample images\n",
    "imagePaths = list(paths.list_images(imagePath))\n",
    "#print('point a')\n",
    "f = open(\"filesize.txt\", \"w\")\n",
    "data = []\n",
    "labels = []\n",
    "\n",
    "count = 1\n",
    "filesize=[]\n",
    "# read all X-Rays in the specified path, and resize them all to 256x256\n",
    "for imagePath in imagePaths:\n",
    "    #print(count)\n",
    "    label = imagePath.split(os.path.sep)[-2]\n",
    "    image = cv2.imread(imagePath)\n",
    "    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
    "    height, width, channels = image.shape\n",
    "    #print (label,' ',height,' ', width,' ', channels)\n",
    "    #f.write(label,' ',height,' ', width,' ', channels)\n",
    "    filesize.append([label,height,width,channels,height*width*channels])\n",
    "    image = cv2.resize(image, (imagesize, imagesize))\n",
    "    data.append(image)\n",
    "    labels.append(label)\n",
    "    count=count+1\n",
    "#print('point b')    \n",
    "#normalise pixel values to real numbers between 0.0 - 1.0\n",
    "data = np.array(data) / 255.0\n",
    "labels = np.array(labels)\n",
    "#print('point c')\n",
    "# perform one-hot encoding for a multi-class labeling\n",
    "label_encoder = LabelEncoder()\n",
    "integer_encoded = label_encoder.fit_transform(labels)\n",
    "labels = to_categorical(integer_encoded)\n",
    "#print('point d')\n",
    "print(\"... ... \", len(data), \"images loaded in multiple classes:\")\n",
    "print(label_encoder.classes_)\n",
    "#print('point e')\n",
    "f.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "calculating the image size in training dataset and saving in a training_image_size.csv file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>height</th>\n",
       "      <th>width</th>\n",
       "      <th>channels</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./newtrain/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./newtrain/covid</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./newtrain/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./newtrain/covid</td>\n",
       "      <td>2539</td>\n",
       "      <td>3050</td>\n",
       "      <td>3</td>\n",
       "      <td>23231850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./newtrain/covid</td>\n",
       "      <td>1330</td>\n",
       "      <td>1796</td>\n",
       "      <td>3</td>\n",
       "      <td>7166040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>./newtrain/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>./newtrain/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89</th>\n",
       "      <td>./newtrain/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>./newtrain/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>./newtrain/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>92 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                   label  height  width  channels      size\n",
       "0       ./newtrain/covid     512    512         3    786432\n",
       "1       ./newtrain/covid    1024   1024         3   3145728\n",
       "2       ./newtrain/covid     512    512         3    786432\n",
       "3       ./newtrain/covid    2539   3050         3  23231850\n",
       "4       ./newtrain/covid    1330   1796         3   7166040\n",
       "..                   ...     ...    ...       ...       ...\n",
       "87  ./newtrain/pneumonia    1024   1024         3   3145728\n",
       "88  ./newtrain/pneumonia    1024   1024         3   3145728\n",
       "89  ./newtrain/pneumonia    1024   1024         3   3145728\n",
       "90  ./newtrain/pneumonia    1024   1024         3   3145728\n",
       "91  ./newtrain/pneumonia    1024   1024         3   3145728\n",
       "\n",
       "[92 rows x 5 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.DataFrame(filesize,columns=[\"label\",\"height\",\"width\",\"channels\",\"size\"])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label                 height  width  channels  size    \n",
       "./newtrain/normal     1024    1024   3         3145728     37\n",
       "./newtrain/pneumonia  1024    1024   3         3145728     22\n",
       "./newtrain/covid      512     512    3         786432       8\n",
       "                      1024    1024   3         3145728      7\n",
       "                      2000    2000   3         12000000     3\n",
       "                      554     561    3         932382       1\n",
       "                      597     737    3         1319967      1\n",
       "                      610     586    3         1072380      1\n",
       "                      695     849    3         1770165      1\n",
       "                      734     895    3         1970790      1\n",
       "                      992     1024   3         3047424      1\n",
       "                      366     446    3         489708       1\n",
       "                      1282    1210   3         4653660      1\n",
       "                      1330    1796   3         7166040      1\n",
       "                      1711    2038   3         10461054     1\n",
       "                      1828    2200   3         12064800     1\n",
       "                      1971    2498   3         14770674     1\n",
       "                      2539    3050   3         23231850     1\n",
       "                      462     497    3         688842       1\n",
       "                      277     375    3         311625       1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.DataFrame(df.value_counts()).to_csv('training_image_size.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['label', 'height', 'width', 'channels', 'size', '0']\n",
      "['./newtrain/normal', '1024', '1024', '3', '3145728', '37']\n",
      "['./newtrain/pneumonia', '1024', '1024', '3', '3145728', '22']\n",
      "['./newtrain/covid', '512', '512', '3', '786432', '8']\n",
      "['./newtrain/covid', '1024', '1024', '3', '3145728', '7']\n",
      "['./newtrain/covid', '2000', '2000', '3', '12000000', '3']\n",
      "['./newtrain/covid', '554', '561', '3', '932382', '1']\n",
      "['./newtrain/covid', '597', '737', '3', '1319967', '1']\n",
      "['./newtrain/covid', '610', '586', '3', '1072380', '1']\n",
      "['./newtrain/covid', '695', '849', '3', '1770165', '1']\n",
      "['./newtrain/covid', '734', '895', '3', '1970790', '1']\n",
      "['./newtrain/covid', '992', '1024', '3', '3047424', '1']\n",
      "['./newtrain/covid', '366', '446', '3', '489708', '1']\n",
      "['./newtrain/covid', '1282', '1210', '3', '4653660', '1']\n",
      "['./newtrain/covid', '1330', '1796', '3', '7166040', '1']\n",
      "['./newtrain/covid', '1711', '2038', '3', '10461054', '1']\n",
      "['./newtrain/covid', '1828', '2200', '3', '12064800', '1']\n",
      "['./newtrain/covid', '1971', '2498', '3', '14770674', '1']\n",
      "['./newtrain/covid', '2539', '3050', '3', '23231850', '1']\n",
      "['./newtrain/covid', '462', '497', '3', '688842', '1']\n",
      "['./newtrain/covid', '277', '375', '3', '311625', '1']\n"
     ]
    }
   ],
   "source": [
    "import csv\n",
    "with open('training_image_size.csv', newline='') as f:\n",
    "    reader = csv.reader(f)\n",
    "    for row in reader:\n",
    "        print(row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label</th>\n",
       "      <th>height</th>\n",
       "      <th>width</th>\n",
       "      <th>channels</th>\n",
       "      <th>size</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>./newtrain/normal</th>\n",
       "      <th>1024</th>\n",
       "      <th>1024</th>\n",
       "      <th>3</th>\n",
       "      <th>3145728</th>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>./newtrain/pneumonia</th>\n",
       "      <th>1024</th>\n",
       "      <th>1024</th>\n",
       "      <th>3</th>\n",
       "      <th>3145728</th>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"18\" valign=\"top\">./newtrain/covid</th>\n",
       "      <th>512</th>\n",
       "      <th>512</th>\n",
       "      <th>3</th>\n",
       "      <th>786432</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <th>1024</th>\n",
       "      <th>3</th>\n",
       "      <th>3145728</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <th>2000</th>\n",
       "      <th>3</th>\n",
       "      <th>12000000</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>554</th>\n",
       "      <th>561</th>\n",
       "      <th>3</th>\n",
       "      <th>932382</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>597</th>\n",
       "      <th>737</th>\n",
       "      <th>3</th>\n",
       "      <th>1319967</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>610</th>\n",
       "      <th>586</th>\n",
       "      <th>3</th>\n",
       "      <th>1072380</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>695</th>\n",
       "      <th>849</th>\n",
       "      <th>3</th>\n",
       "      <th>1770165</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>734</th>\n",
       "      <th>895</th>\n",
       "      <th>3</th>\n",
       "      <th>1970790</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>992</th>\n",
       "      <th>1024</th>\n",
       "      <th>3</th>\n",
       "      <th>3047424</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366</th>\n",
       "      <th>446</th>\n",
       "      <th>3</th>\n",
       "      <th>489708</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1282</th>\n",
       "      <th>1210</th>\n",
       "      <th>3</th>\n",
       "      <th>4653660</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1330</th>\n",
       "      <th>1796</th>\n",
       "      <th>3</th>\n",
       "      <th>7166040</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1711</th>\n",
       "      <th>2038</th>\n",
       "      <th>3</th>\n",
       "      <th>10461054</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1828</th>\n",
       "      <th>2200</th>\n",
       "      <th>3</th>\n",
       "      <th>12064800</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1971</th>\n",
       "      <th>2498</th>\n",
       "      <th>3</th>\n",
       "      <th>14770674</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2539</th>\n",
       "      <th>3050</th>\n",
       "      <th>3</th>\n",
       "      <th>23231850</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>462</th>\n",
       "      <th>497</th>\n",
       "      <th>3</th>\n",
       "      <th>688842</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277</th>\n",
       "      <th>375</th>\n",
       "      <th>3</th>\n",
       "      <th>311625</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      0\n",
       "label                height width channels size        \n",
       "./newtrain/normal    1024   1024  3        3145728   37\n",
       "./newtrain/pneumonia 1024   1024  3        3145728   22\n",
       "./newtrain/covid     512    512   3        786432     8\n",
       "                     1024   1024  3        3145728    7\n",
       "                     2000   2000  3        12000000   3\n",
       "                     554    561   3        932382     1\n",
       "                     597    737   3        1319967    1\n",
       "                     610    586   3        1072380    1\n",
       "                     695    849   3        1770165    1\n",
       "                     734    895   3        1970790    1\n",
       "                     992    1024  3        3047424    1\n",
       "                     366    446   3        489708     1\n",
       "                     1282   1210  3        4653660    1\n",
       "                     1330   1796  3        7166040    1\n",
       "                     1711   2038  3        10461054   1\n",
       "                     1828   2200  3        12064800   1\n",
       "                     1971   2498  3        14770674   1\n",
       "                     2539   3050  3        23231850   1\n",
       "                     462    497   3        688842     1\n",
       "                     277    375   3        311625     1"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(df.value_counts()).head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['./newtrain/covid' './newtrain/normal' './newtrain/pneumonia']\n",
      "\n",
      "(3,)\n",
      "\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.str_'>\n"
     ]
    }
   ],
   "source": [
    "classes_of_data=label_encoder.classes_\n",
    "\n",
    "print(classes_of_data)\n",
    "print()\n",
    "print(classes_of_data.shape)\n",
    "print()\n",
    "print(type(classes_of_data))\n",
    "print(type(classes_of_data[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./newtrain/covid\n",
      "./newtrain/normal\n",
      "./newtrain/pneumonia\n",
      "['./newtrain/covid', './newtrain/normal', './newtrain/pneumonia']\n",
      "['covid', 'normal', 'pneumonia']\n"
     ]
    }
   ],
   "source": [
    "c1=[]\n",
    "c2=[]\n",
    "for i in range(len(classes_of_data)):\n",
    "    print(classes_of_data[i])\n",
    "    c1.append(classes_of_data[i])\n",
    "    c2.append(re.sub(r'^./newtrain/','',classes_of_data[i]))\n",
    "\n",
    "print(c1)\n",
    "print(c2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(classes_of_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'list'>\n",
      "<class 'str'>\n"
     ]
    }
   ],
   "source": [
    "print(type(c2))\n",
    "print(type(c2[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./newtrain/covid covid\n",
      "./newtrain/normal normal\n",
      "./newtrain/pneumonia pneumonia\n"
     ]
    }
   ],
   "source": [
    "for i in (classes_of_data):\n",
    "    print(i,re.sub(r'^./newtrain/','',i))\n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(92, 256, 256, 3)\n"
     ]
    }
   ],
   "source": [
    "print(data.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(imagePath)\n",
    "#print((imagePaths))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading sample image"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Loading sample image from training data set: the first image from covid, normal (healthy), and pneumonia class are loaded as a sample image, which is used in the research paper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load and display an image with Matplotlib\n",
    "from matplotlib import image\n",
    "from matplotlib import pyplot\n",
    "import cv2\n",
    "cov = image.imread('./newtrain/covid\\\\cov_1110.jpg')\n",
    "nor = image.imread('./newtrain/normal\\\\normal-0000.jpg')\n",
    "pne = image.imread('./newtrain/pneumonia\\\\pneumonia0000.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uint8\n",
      "(512, 512, 3)\n",
      "\n",
      "uint8\n",
      "(1024, 1024)\n",
      "\n",
      "uint8\n",
      "(1024, 1024)\n"
     ]
    }
   ],
   "source": [
    "# summarize shape of the pixel array\n",
    "print(cov.dtype)\n",
    "print(cov.shape)\n",
    "print()\n",
    "print(nor.dtype)\n",
    "print(nor.shape)\n",
    "print()\n",
    "print(pne.dtype)\n",
    "print(pne.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 960x800 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.pyplot import figure\n",
    "\n",
    "figure(figsize=(12, 10), dpi=80)\n",
    "# plot all three images using matplotlib\n",
    "pyplot.subplot(131)\n",
    "pyplot.imshow(cov,cmap='gray')\n",
    "plt.title(\"COVID\")\n",
    "pyplot.subplot(132)\n",
    "pyplot.imshow(nor,cmap='gray')\n",
    "plt.title(\"Normal\")\n",
    "pyplot.subplot(133)\n",
    "pyplot.imshow(pne,cmap='gray')\n",
    "plt.title(\"Pneumonia\")\n",
    "\n",
    "plt.savefig( '{}_Sample_Images.png'.format(dt))\n",
    "plt.savefig(\"sample_images.pdf\")\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./newtrain/covid     ....0\n",
      "./newtrain/normal    ....1\n",
      "./newtrain/pneumonia ....2\n"
     ]
    }
   ],
   "source": [
    "list_of_classes=list(label_encoder.classes_)\n",
    "#print(label_encoder.transform(list_of_classes))\n",
    "for i ,j in zip(list_of_classes,label_encoder.transform(list_of_classes)):\n",
    "    print('{:20}'.format(i),'{:.>5}'.format(j))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n"
     ]
    }
   ],
   "source": [
    "print(len(list_of_classes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./newtrain/covid     .......covid ....0\n",
      "./newtrain/normal    ......normal ....1\n",
      "./newtrain/pneumonia ...pneumonia ....2\n"
     ]
    }
   ],
   "source": [
    "for i ,j, k in zip(list_of_classes,label_encoder.transform(list_of_classes),c2):\n",
    "    print('{:20}'.format(i),'{:.>12}'.format(k),'{:.>5}'.format(j))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(92, 3)\n",
      "[[1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]]\n"
     ]
    }
   ],
   "source": [
    "print(labels.shape)\n",
    "print(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " size of data:  18087936 \n",
      " No of rows:  92 \n",
      " No of labels:  92 \n",
      " Type:  <class 'numpy.ndarray'> \n",
      " rows:  92 \n",
      " columns:  256 \n",
      " total_length:  23552\n"
     ]
    }
   ],
   "source": [
    "print(' size of data: ',data.size, '\\n No of rows: ',len(data),'\\n No of labels: ',len(labels),'\\n Type: ',type(data),'\\n','rows: ',len(data),'\\n','columns: ',len(data[0]),'\\n','total_length: ',len(data)*len(data[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "92 256 256 3\n"
     ]
    }
   ],
   "source": [
    "print(len(data),len(data[0]),len(data[0][0]),len(data[0][0][0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training Model: Add in basic data augment, re-compose the model, then train it  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trainX: 73\n",
      "testX: 19\n",
      "trainY: 73\n",
      "testY: 19\n",
      "Compiling model...\n",
      "Training the full stack model...\n",
      "Epoch 1/2\n",
      "4/4 [==============================] - 1895s 552s/step - loss: 1.6974 - accuracy: 0.4103 - val_loss: 1.1607 - val_accuracy: 0.4211ETA: 0s - loss: 1.6826 - accuracy: 0.4208\n",
      "\n",
      "Epoch 00001: val_accuracy improved from -inf to 0.42105, saving model to best_model.h5\n",
      "Epoch 2/2\n",
      "4/4 [==============================] - 515s 84s/step - loss: 1.6570 - accuracy: 0.4297 - val_loss: 1.1236 - val_accuracy: 0.3684\n",
      "\n",
      "Epoch 00002: val_accuracy did not improve from 0.42105\n"
     ]
    }
   ],
   "source": [
    "#baseModel3 = VGG16(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(imagesize, imagesize, 3)))\n",
    "#baseModel3 = InceptionV3(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(imagesize, imagesize, 3)))\n",
    "#baseModel = VGG16(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(imagesize, imagesize, 3)))\n",
    "\n",
    "#baseModel3 = ResNet50(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(imagesize, imagesize, 3)))\n",
    "baseModel3 = DenseNet121(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(imagesize, imagesize, 3)))\n",
    "#baseModel = VGG16(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(256, 256, 3)))\n",
    "#baseModel = ResNet50(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(256, 256, 3)))\n",
    "\n",
    "# split the data between train and validation.\n",
    "(trainX, testX, trainY, testY) = train_test_split(data, labels, test_size=0.20, stratify=labels, random_state=42)\n",
    "\n",
    "print(\"trainX:\",trainX.shape[0])\n",
    "print(\"testX:\",testX.shape[0])\n",
    "print(\"trainY:\",trainY.shape[0])\n",
    "print(\"testY:\",testY.shape[0])\n",
    "\n",
    "# add on a simple Augmentation. Note: too many Augumentation doesn't actually help in this case - I found during the test. \n",
    "trainAug = ImageDataGenerator(rotation_range=15, fill_mode=\"nearest\")\n",
    "\n",
    "\n",
    "#baseModel3 = DenseNet121(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(imagesize, imagesize, 3)))\n",
    "#Add on a couple of custom CNN layers on top of the Inception V3 model. \n",
    "headModel3 = baseModel3.output\n",
    "#tf.keras.layers.\n",
    "#headModel3 = GlobalAveragePooling2D()(headModel3)\n",
    "headModel3 = AveragePooling2D(pool_size=(4, 4))(headModel3)\n",
    "headModel3 = Flatten(name=\"flatten\")(headModel3)\n",
    "headModel3 = Dense(64, activation=\"relu\")(headModel3)\n",
    "headModel3 = Dropout(0.5)(headModel3)\n",
    "headModel3 = Dense(3, activation=\"softmax\")(headModel3) # as we have 3 classes: covid, normal and pneumonia \n",
    "\n",
    "# Compose the final model\n",
    "model = Model(inputs=baseModel3.input, \n",
    "              outputs=headModel3)\n",
    "\n",
    "# Unfreeze pre-trained Inception \"ImageNet\" weights for re-training since I got a Navidia T4 GPU to play with anyway\n",
    "#for layer in baseModel.layers:\n",
    "#    layer.trainable = False\n",
    "\n",
    "print(\"Compiling model...\")\n",
    "\n",
    "opt = Adam(lr=INIT_LR, \n",
    "           decay=INIT_LR / EPOCHS)\n",
    "\n",
    "model.compile(loss=\"categorical_crossentropy\", \n",
    "              optimizer=opt, \n",
    "              metrics=[\"accuracy\"])\n",
    "\n",
    "# pip3 install pydot-ng\n",
    "# pip3 install graphviz\n",
    "#\n",
    "#plot_model(model, to_file='model_plot.png', show_shapes=True, show_layer_names=True)\n",
    "\n",
    "# simple early stopping\n",
    "# We can update the example and add very simple early stopping.\n",
    "# As soon as the loss of the model begins to increase on the test dataset, we will stop training.\n",
    "# monitor: allows you to specify the performance measure to monitor in order to end training.\n",
    "# mode: we would seek a maximum for validation accuracy.\n",
    "# verbose = 1: Once stopped, the callback will print the epoch number.\n",
    "# min_delta: for 0.5% for accuracy. \n",
    "# baseline: \n",
    "#es = tf.keras.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', verbose=1, min_delta=0.5)#, baseline=0.4)\n",
    "#from tensorflow.keras.callbacks import ReduceLROnPlateau\n",
    "\n",
    "es = tf.keras.callbacks.ReduceLROnPlateau(\n",
    "        monitor='val_accuracy', # monitor: quantity to be monitored.\n",
    "        factor=0.001, # factor: factor by which the learning rate will be reduced. new_lr = lr * factor.\n",
    "        patience=30, # patience: number of epochs with no improvement after which learning rate will be reduced.\n",
    "        mode='auto', # mode: one of {'auto', 'min', 'max'}  in 'auto' mode, the direction is automatically inferred \n",
    "        #                                                                   from the name of the monitored quantity.\n",
    "        min_delta=0.05, # min_delta: threshold for measuring the new optimum, to only focus on significant changes; \n",
    "        cooldown=5, # cooldown: number of epochs to wait before resuming normal operation after lr has been reduced.\n",
    "        min_lr=INIT_LR, # min_lr: lower bound on the learning rate.\n",
    "        verbose=1, # displays epoch number\n",
    "    )\n",
    "\n",
    "mc = tf.keras.callbacks.ModelCheckpoint('best_model.h5', \n",
    "                                        monitor='val_accuracy', \n",
    "                                        mode='max', \n",
    "                                        verbose=1, \n",
    "                                        save_best_only=True)\n",
    "\n",
    "# We can improve the trigger for early stopping by waiting a while before stopping.\n",
    "# patient early stopping: In this case, we will wait 200 epochs before training is stopped. \n",
    "#          Specifically, this means that we will allow training to continue for up to an additional 5 epochs \n",
    "#          after the point that validation loss started to degrade, giving the training process an opportunity \n",
    "#          to get across flat spots or find some additional improvement.\n",
    "# es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=5)\n",
    "\n",
    "# train the full model, since we unfroze the pre-trained weights above\n",
    "print(\"Training the full stack model...\")\n",
    "H = model.fit( trainAug.flow(trainX, trainY, batch_size=BS), \n",
    "                        steps_per_epoch=len(trainX) // (BS),\n",
    "                        validation_data=(testX, testY), \n",
    "                        validation_steps=len(testX) // (BS), \n",
    "                        epochs=EPOCHS,\n",
    "                        callbacks=[es,mc]                        \n",
    "                       )\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 256, 256, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d (ZeroPadding2D)  (None, 262, 262, 3)  0           input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv1/conv (Conv2D)             (None, 128, 128, 64) 9408        zero_padding2d[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv1/bn (BatchNormalization)   (None, 128, 128, 64) 256         conv1/conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv1/relu (Activation)         (None, 128, 128, 64) 0           conv1/bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d_1 (ZeroPadding2D (None, 130, 130, 64) 0           conv1/relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool1 (MaxPooling2D)            (None, 64, 64, 64)   0           zero_padding2d_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_0_bn (BatchNormali (None, 64, 64, 64)   256         pool1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_0_relu (Activation (None, 64, 64, 64)   0           conv2_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_conv (Conv2D)    (None, 64, 64, 128)  8192        conv2_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_relu (Activation (None, 64, 64, 128)  0           conv2_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_2_conv (Conv2D)    (None, 64, 64, 32)   36864       conv2_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_concat (Concatenat (None, 64, 64, 96)   0           pool1[0][0]                      \n",
      "                                                                 conv2_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_0_bn (BatchNormali (None, 64, 64, 96)   384         conv2_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_0_relu (Activation (None, 64, 64, 96)   0           conv2_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_conv (Conv2D)    (None, 64, 64, 128)  12288       conv2_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_relu (Activation (None, 64, 64, 128)  0           conv2_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_2_conv (Conv2D)    (None, 64, 64, 32)   36864       conv2_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_concat (Concatenat (None, 64, 64, 128)  0           conv2_block1_concat[0][0]        \n",
      "                                                                 conv2_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_0_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_0_relu (Activation (None, 64, 64, 128)  0           conv2_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_conv (Conv2D)    (None, 64, 64, 128)  16384       conv2_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_relu (Activation (None, 64, 64, 128)  0           conv2_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_2_conv (Conv2D)    (None, 64, 64, 32)   36864       conv2_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_concat (Concatenat (None, 64, 64, 160)  0           conv2_block2_concat[0][0]        \n",
      "                                                                 conv2_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_0_bn (BatchNormali (None, 64, 64, 160)  640         conv2_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_0_relu (Activation (None, 64, 64, 160)  0           conv2_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_conv (Conv2D)    (None, 64, 64, 128)  20480       conv2_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_relu (Activation (None, 64, 64, 128)  0           conv2_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_2_conv (Conv2D)    (None, 64, 64, 32)   36864       conv2_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_concat (Concatenat (None, 64, 64, 192)  0           conv2_block3_concat[0][0]        \n",
      "                                                                 conv2_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_0_bn (BatchNormali (None, 64, 64, 192)  768         conv2_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_0_relu (Activation (None, 64, 64, 192)  0           conv2_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_conv (Conv2D)    (None, 64, 64, 128)  24576       conv2_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_relu (Activation (None, 64, 64, 128)  0           conv2_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_2_conv (Conv2D)    (None, 64, 64, 32)   36864       conv2_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_concat (Concatenat (None, 64, 64, 224)  0           conv2_block4_concat[0][0]        \n",
      "                                                                 conv2_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_0_bn (BatchNormali (None, 64, 64, 224)  896         conv2_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_0_relu (Activation (None, 64, 64, 224)  0           conv2_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_conv (Conv2D)    (None, 64, 64, 128)  28672       conv2_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_bn (BatchNormali (None, 64, 64, 128)  512         conv2_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_relu (Activation (None, 64, 64, 128)  0           conv2_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_2_conv (Conv2D)    (None, 64, 64, 32)   36864       conv2_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_concat (Concatenat (None, 64, 64, 256)  0           conv2_block5_concat[0][0]        \n",
      "                                                                 conv2_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "pool2_bn (BatchNormalization)   (None, 64, 64, 256)  1024        conv2_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "pool2_relu (Activation)         (None, 64, 64, 256)  0           pool2_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool2_conv (Conv2D)             (None, 64, 64, 128)  32768       pool2_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool2_pool (AveragePooling2D)   (None, 32, 32, 128)  0           pool2_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_0_bn (BatchNormali (None, 32, 32, 128)  512         pool2_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_0_relu (Activation (None, 32, 32, 128)  0           conv3_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_conv (Conv2D)    (None, 32, 32, 128)  16384       conv3_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_relu (Activation (None, 32, 32, 128)  0           conv3_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_concat (Concatenat (None, 32, 32, 160)  0           pool2_pool[0][0]                 \n",
      "                                                                 conv3_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_0_bn (BatchNormali (None, 32, 32, 160)  640         conv3_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_0_relu (Activation (None, 32, 32, 160)  0           conv3_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_conv (Conv2D)    (None, 32, 32, 128)  20480       conv3_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_relu (Activation (None, 32, 32, 128)  0           conv3_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_concat (Concatenat (None, 32, 32, 192)  0           conv3_block1_concat[0][0]        \n",
      "                                                                 conv3_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_0_bn (BatchNormali (None, 32, 32, 192)  768         conv3_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_0_relu (Activation (None, 32, 32, 192)  0           conv3_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_conv (Conv2D)    (None, 32, 32, 128)  24576       conv3_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_relu (Activation (None, 32, 32, 128)  0           conv3_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_concat (Concatenat (None, 32, 32, 224)  0           conv3_block2_concat[0][0]        \n",
      "                                                                 conv3_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_0_bn (BatchNormali (None, 32, 32, 224)  896         conv3_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_0_relu (Activation (None, 32, 32, 224)  0           conv3_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_conv (Conv2D)    (None, 32, 32, 128)  28672       conv3_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_relu (Activation (None, 32, 32, 128)  0           conv3_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_concat (Concatenat (None, 32, 32, 256)  0           conv3_block3_concat[0][0]        \n",
      "                                                                 conv3_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_0_bn (BatchNormali (None, 32, 32, 256)  1024        conv3_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_0_relu (Activation (None, 32, 32, 256)  0           conv3_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_conv (Conv2D)    (None, 32, 32, 128)  32768       conv3_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_relu (Activation (None, 32, 32, 128)  0           conv3_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_concat (Concatenat (None, 32, 32, 288)  0           conv3_block4_concat[0][0]        \n",
      "                                                                 conv3_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_0_bn (BatchNormali (None, 32, 32, 288)  1152        conv3_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_0_relu (Activation (None, 32, 32, 288)  0           conv3_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_conv (Conv2D)    (None, 32, 32, 128)  36864       conv3_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_relu (Activation (None, 32, 32, 128)  0           conv3_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_concat (Concatenat (None, 32, 32, 320)  0           conv3_block5_concat[0][0]        \n",
      "                                                                 conv3_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_0_bn (BatchNormali (None, 32, 32, 320)  1280        conv3_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_0_relu (Activation (None, 32, 32, 320)  0           conv3_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_conv (Conv2D)    (None, 32, 32, 128)  40960       conv3_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_relu (Activation (None, 32, 32, 128)  0           conv3_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_concat (Concatenat (None, 32, 32, 352)  0           conv3_block6_concat[0][0]        \n",
      "                                                                 conv3_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_0_bn (BatchNormali (None, 32, 32, 352)  1408        conv3_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_0_relu (Activation (None, 32, 32, 352)  0           conv3_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_conv (Conv2D)    (None, 32, 32, 128)  45056       conv3_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_relu (Activation (None, 32, 32, 128)  0           conv3_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_concat (Concatenat (None, 32, 32, 384)  0           conv3_block7_concat[0][0]        \n",
      "                                                                 conv3_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_0_bn (BatchNormali (None, 32, 32, 384)  1536        conv3_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_0_relu (Activation (None, 32, 32, 384)  0           conv3_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_conv (Conv2D)    (None, 32, 32, 128)  49152       conv3_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_bn (BatchNormali (None, 32, 32, 128)  512         conv3_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_relu (Activation (None, 32, 32, 128)  0           conv3_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_2_conv (Conv2D)    (None, 32, 32, 32)   36864       conv3_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_concat (Concatenat (None, 32, 32, 416)  0           conv3_block8_concat[0][0]        \n",
      "                                                                 conv3_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_0_bn (BatchNormal (None, 32, 32, 416)  1664        conv3_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_0_relu (Activatio (None, 32, 32, 416)  0           conv3_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_conv (Conv2D)   (None, 32, 32, 128)  53248       conv3_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_bn (BatchNormal (None, 32, 32, 128)  512         conv3_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_relu (Activatio (None, 32, 32, 128)  0           conv3_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_2_conv (Conv2D)   (None, 32, 32, 32)   36864       conv3_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_concat (Concatena (None, 32, 32, 448)  0           conv3_block9_concat[0][0]        \n",
      "                                                                 conv3_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_0_bn (BatchNormal (None, 32, 32, 448)  1792        conv3_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_0_relu (Activatio (None, 32, 32, 448)  0           conv3_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_conv (Conv2D)   (None, 32, 32, 128)  57344       conv3_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_bn (BatchNormal (None, 32, 32, 128)  512         conv3_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_relu (Activatio (None, 32, 32, 128)  0           conv3_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_2_conv (Conv2D)   (None, 32, 32, 32)   36864       conv3_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_concat (Concatena (None, 32, 32, 480)  0           conv3_block10_concat[0][0]       \n",
      "                                                                 conv3_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_0_bn (BatchNormal (None, 32, 32, 480)  1920        conv3_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_0_relu (Activatio (None, 32, 32, 480)  0           conv3_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_conv (Conv2D)   (None, 32, 32, 128)  61440       conv3_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_bn (BatchNormal (None, 32, 32, 128)  512         conv3_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_relu (Activatio (None, 32, 32, 128)  0           conv3_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_2_conv (Conv2D)   (None, 32, 32, 32)   36864       conv3_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_concat (Concatena (None, 32, 32, 512)  0           conv3_block11_concat[0][0]       \n",
      "                                                                 conv3_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool3_bn (BatchNormalization)   (None, 32, 32, 512)  2048        conv3_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool3_relu (Activation)         (None, 32, 32, 512)  0           pool3_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool3_conv (Conv2D)             (None, 32, 32, 256)  131072      pool3_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool3_pool (AveragePooling2D)   (None, 16, 16, 256)  0           pool3_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_0_bn (BatchNormali (None, 16, 16, 256)  1024        pool3_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_0_relu (Activation (None, 16, 16, 256)  0           conv4_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_conv (Conv2D)    (None, 16, 16, 128)  32768       conv4_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_relu (Activation (None, 16, 16, 128)  0           conv4_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_concat (Concatenat (None, 16, 16, 288)  0           pool3_pool[0][0]                 \n",
      "                                                                 conv4_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_0_bn (BatchNormali (None, 16, 16, 288)  1152        conv4_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_0_relu (Activation (None, 16, 16, 288)  0           conv4_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_conv (Conv2D)    (None, 16, 16, 128)  36864       conv4_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_relu (Activation (None, 16, 16, 128)  0           conv4_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_concat (Concatenat (None, 16, 16, 320)  0           conv4_block1_concat[0][0]        \n",
      "                                                                 conv4_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_0_bn (BatchNormali (None, 16, 16, 320)  1280        conv4_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_0_relu (Activation (None, 16, 16, 320)  0           conv4_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_conv (Conv2D)    (None, 16, 16, 128)  40960       conv4_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_relu (Activation (None, 16, 16, 128)  0           conv4_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_concat (Concatenat (None, 16, 16, 352)  0           conv4_block2_concat[0][0]        \n",
      "                                                                 conv4_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_0_bn (BatchNormali (None, 16, 16, 352)  1408        conv4_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_0_relu (Activation (None, 16, 16, 352)  0           conv4_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_conv (Conv2D)    (None, 16, 16, 128)  45056       conv4_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_relu (Activation (None, 16, 16, 128)  0           conv4_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_concat (Concatenat (None, 16, 16, 384)  0           conv4_block3_concat[0][0]        \n",
      "                                                                 conv4_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_0_bn (BatchNormali (None, 16, 16, 384)  1536        conv4_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_0_relu (Activation (None, 16, 16, 384)  0           conv4_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_conv (Conv2D)    (None, 16, 16, 128)  49152       conv4_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_relu (Activation (None, 16, 16, 128)  0           conv4_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_concat (Concatenat (None, 16, 16, 416)  0           conv4_block4_concat[0][0]        \n",
      "                                                                 conv4_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_0_bn (BatchNormali (None, 16, 16, 416)  1664        conv4_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_0_relu (Activation (None, 16, 16, 416)  0           conv4_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_conv (Conv2D)    (None, 16, 16, 128)  53248       conv4_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_relu (Activation (None, 16, 16, 128)  0           conv4_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_concat (Concatenat (None, 16, 16, 448)  0           conv4_block5_concat[0][0]        \n",
      "                                                                 conv4_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_0_bn (BatchNormali (None, 16, 16, 448)  1792        conv4_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_0_relu (Activation (None, 16, 16, 448)  0           conv4_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_conv (Conv2D)    (None, 16, 16, 128)  57344       conv4_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_relu (Activation (None, 16, 16, 128)  0           conv4_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_concat (Concatenat (None, 16, 16, 480)  0           conv4_block6_concat[0][0]        \n",
      "                                                                 conv4_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_0_bn (BatchNormali (None, 16, 16, 480)  1920        conv4_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_0_relu (Activation (None, 16, 16, 480)  0           conv4_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_conv (Conv2D)    (None, 16, 16, 128)  61440       conv4_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_relu (Activation (None, 16, 16, 128)  0           conv4_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_concat (Concatenat (None, 16, 16, 512)  0           conv4_block7_concat[0][0]        \n",
      "                                                                 conv4_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_0_bn (BatchNormali (None, 16, 16, 512)  2048        conv4_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_0_relu (Activation (None, 16, 16, 512)  0           conv4_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_conv (Conv2D)    (None, 16, 16, 128)  65536       conv4_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_bn (BatchNormali (None, 16, 16, 128)  512         conv4_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_relu (Activation (None, 16, 16, 128)  0           conv4_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_2_conv (Conv2D)    (None, 16, 16, 32)   36864       conv4_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_concat (Concatenat (None, 16, 16, 544)  0           conv4_block8_concat[0][0]        \n",
      "                                                                 conv4_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_0_bn (BatchNormal (None, 16, 16, 544)  2176        conv4_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_0_relu (Activatio (None, 16, 16, 544)  0           conv4_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_conv (Conv2D)   (None, 16, 16, 128)  69632       conv4_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_concat (Concatena (None, 16, 16, 576)  0           conv4_block9_concat[0][0]        \n",
      "                                                                 conv4_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_0_bn (BatchNormal (None, 16, 16, 576)  2304        conv4_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_0_relu (Activatio (None, 16, 16, 576)  0           conv4_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_conv (Conv2D)   (None, 16, 16, 128)  73728       conv4_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_concat (Concatena (None, 16, 16, 608)  0           conv4_block10_concat[0][0]       \n",
      "                                                                 conv4_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_0_bn (BatchNormal (None, 16, 16, 608)  2432        conv4_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_0_relu (Activatio (None, 16, 16, 608)  0           conv4_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_conv (Conv2D)   (None, 16, 16, 128)  77824       conv4_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_concat (Concatena (None, 16, 16, 640)  0           conv4_block11_concat[0][0]       \n",
      "                                                                 conv4_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_0_bn (BatchNormal (None, 16, 16, 640)  2560        conv4_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_0_relu (Activatio (None, 16, 16, 640)  0           conv4_block13_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_conv (Conv2D)   (None, 16, 16, 128)  81920       conv4_block13_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block13_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block13_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block13_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_concat (Concatena (None, 16, 16, 672)  0           conv4_block12_concat[0][0]       \n",
      "                                                                 conv4_block13_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_0_bn (BatchNormal (None, 16, 16, 672)  2688        conv4_block13_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_0_relu (Activatio (None, 16, 16, 672)  0           conv4_block14_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_conv (Conv2D)   (None, 16, 16, 128)  86016       conv4_block14_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block14_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block14_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block14_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_concat (Concatena (None, 16, 16, 704)  0           conv4_block13_concat[0][0]       \n",
      "                                                                 conv4_block14_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_0_bn (BatchNormal (None, 16, 16, 704)  2816        conv4_block14_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_0_relu (Activatio (None, 16, 16, 704)  0           conv4_block15_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_conv (Conv2D)   (None, 16, 16, 128)  90112       conv4_block15_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block15_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block15_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block15_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_concat (Concatena (None, 16, 16, 736)  0           conv4_block14_concat[0][0]       \n",
      "                                                                 conv4_block15_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_0_bn (BatchNormal (None, 16, 16, 736)  2944        conv4_block15_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_0_relu (Activatio (None, 16, 16, 736)  0           conv4_block16_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_conv (Conv2D)   (None, 16, 16, 128)  94208       conv4_block16_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block16_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block16_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block16_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_concat (Concatena (None, 16, 16, 768)  0           conv4_block15_concat[0][0]       \n",
      "                                                                 conv4_block16_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_0_bn (BatchNormal (None, 16, 16, 768)  3072        conv4_block16_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_0_relu (Activatio (None, 16, 16, 768)  0           conv4_block17_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_conv (Conv2D)   (None, 16, 16, 128)  98304       conv4_block17_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block17_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block17_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block17_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_concat (Concatena (None, 16, 16, 800)  0           conv4_block16_concat[0][0]       \n",
      "                                                                 conv4_block17_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_0_bn (BatchNormal (None, 16, 16, 800)  3200        conv4_block17_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_0_relu (Activatio (None, 16, 16, 800)  0           conv4_block18_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_conv (Conv2D)   (None, 16, 16, 128)  102400      conv4_block18_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block18_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block18_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block18_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_concat (Concatena (None, 16, 16, 832)  0           conv4_block17_concat[0][0]       \n",
      "                                                                 conv4_block18_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_0_bn (BatchNormal (None, 16, 16, 832)  3328        conv4_block18_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_0_relu (Activatio (None, 16, 16, 832)  0           conv4_block19_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_conv (Conv2D)   (None, 16, 16, 128)  106496      conv4_block19_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block19_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block19_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block19_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_concat (Concatena (None, 16, 16, 864)  0           conv4_block18_concat[0][0]       \n",
      "                                                                 conv4_block19_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_0_bn (BatchNormal (None, 16, 16, 864)  3456        conv4_block19_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_0_relu (Activatio (None, 16, 16, 864)  0           conv4_block20_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_conv (Conv2D)   (None, 16, 16, 128)  110592      conv4_block20_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block20_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block20_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block20_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_concat (Concatena (None, 16, 16, 896)  0           conv4_block19_concat[0][0]       \n",
      "                                                                 conv4_block20_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_0_bn (BatchNormal (None, 16, 16, 896)  3584        conv4_block20_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_0_relu (Activatio (None, 16, 16, 896)  0           conv4_block21_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_conv (Conv2D)   (None, 16, 16, 128)  114688      conv4_block21_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block21_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block21_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block21_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_concat (Concatena (None, 16, 16, 928)  0           conv4_block20_concat[0][0]       \n",
      "                                                                 conv4_block21_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_0_bn (BatchNormal (None, 16, 16, 928)  3712        conv4_block21_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_0_relu (Activatio (None, 16, 16, 928)  0           conv4_block22_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_conv (Conv2D)   (None, 16, 16, 128)  118784      conv4_block22_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block22_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block22_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block22_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_concat (Concatena (None, 16, 16, 960)  0           conv4_block21_concat[0][0]       \n",
      "                                                                 conv4_block22_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_0_bn (BatchNormal (None, 16, 16, 960)  3840        conv4_block22_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_0_relu (Activatio (None, 16, 16, 960)  0           conv4_block23_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_conv (Conv2D)   (None, 16, 16, 128)  122880      conv4_block23_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block23_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block23_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block23_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_concat (Concatena (None, 16, 16, 992)  0           conv4_block22_concat[0][0]       \n",
      "                                                                 conv4_block23_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_0_bn (BatchNormal (None, 16, 16, 992)  3968        conv4_block23_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_0_relu (Activatio (None, 16, 16, 992)  0           conv4_block24_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_conv (Conv2D)   (None, 16, 16, 128)  126976      conv4_block24_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_bn (BatchNormal (None, 16, 16, 128)  512         conv4_block24_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_relu (Activatio (None, 16, 16, 128)  0           conv4_block24_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_2_conv (Conv2D)   (None, 16, 16, 32)   36864       conv4_block24_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_concat (Concatena (None, 16, 16, 1024) 0           conv4_block23_concat[0][0]       \n",
      "                                                                 conv4_block24_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool4_bn (BatchNormalization)   (None, 16, 16, 1024) 4096        conv4_block24_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool4_relu (Activation)         (None, 16, 16, 1024) 0           pool4_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool4_conv (Conv2D)             (None, 16, 16, 512)  524288      pool4_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool4_pool (AveragePooling2D)   (None, 8, 8, 512)    0           pool4_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_0_bn (BatchNormali (None, 8, 8, 512)    2048        pool4_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_0_relu (Activation (None, 8, 8, 512)    0           conv5_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_conv (Conv2D)    (None, 8, 8, 128)    65536       conv5_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_relu (Activation (None, 8, 8, 128)    0           conv5_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_concat (Concatenat (None, 8, 8, 544)    0           pool4_pool[0][0]                 \n",
      "                                                                 conv5_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_0_bn (BatchNormali (None, 8, 8, 544)    2176        conv5_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_0_relu (Activation (None, 8, 8, 544)    0           conv5_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_conv (Conv2D)    (None, 8, 8, 128)    69632       conv5_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_relu (Activation (None, 8, 8, 128)    0           conv5_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_concat (Concatenat (None, 8, 8, 576)    0           conv5_block1_concat[0][0]        \n",
      "                                                                 conv5_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_0_bn (BatchNormali (None, 8, 8, 576)    2304        conv5_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_0_relu (Activation (None, 8, 8, 576)    0           conv5_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "conv5_block3_1_conv (Conv2D)    (None, 8, 8, 128)    73728       conv5_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_relu (Activation (None, 8, 8, 128)    0           conv5_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_concat (Concatenat (None, 8, 8, 608)    0           conv5_block2_concat[0][0]        \n",
      "                                                                 conv5_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_0_bn (BatchNormali (None, 8, 8, 608)    2432        conv5_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_0_relu (Activation (None, 8, 8, 608)    0           conv5_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_conv (Conv2D)    (None, 8, 8, 128)    77824       conv5_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_relu (Activation (None, 8, 8, 128)    0           conv5_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_concat (Concatenat (None, 8, 8, 640)    0           conv5_block3_concat[0][0]        \n",
      "                                                                 conv5_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_0_bn (BatchNormali (None, 8, 8, 640)    2560        conv5_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_0_relu (Activation (None, 8, 8, 640)    0           conv5_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_conv (Conv2D)    (None, 8, 8, 128)    81920       conv5_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_relu (Activation (None, 8, 8, 128)    0           conv5_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_concat (Concatenat (None, 8, 8, 672)    0           conv5_block4_concat[0][0]        \n",
      "                                                                 conv5_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_0_bn (BatchNormali (None, 8, 8, 672)    2688        conv5_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_0_relu (Activation (None, 8, 8, 672)    0           conv5_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_conv (Conv2D)    (None, 8, 8, 128)    86016       conv5_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_relu (Activation (None, 8, 8, 128)    0           conv5_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_concat (Concatenat (None, 8, 8, 704)    0           conv5_block5_concat[0][0]        \n",
      "                                                                 conv5_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_0_bn (BatchNormali (None, 8, 8, 704)    2816        conv5_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_0_relu (Activation (None, 8, 8, 704)    0           conv5_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_conv (Conv2D)    (None, 8, 8, 128)    90112       conv5_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_relu (Activation (None, 8, 8, 128)    0           conv5_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_concat (Concatenat (None, 8, 8, 736)    0           conv5_block6_concat[0][0]        \n",
      "                                                                 conv5_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_0_bn (BatchNormali (None, 8, 8, 736)    2944        conv5_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_0_relu (Activation (None, 8, 8, 736)    0           conv5_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_conv (Conv2D)    (None, 8, 8, 128)    94208       conv5_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_relu (Activation (None, 8, 8, 128)    0           conv5_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_concat (Concatenat (None, 8, 8, 768)    0           conv5_block7_concat[0][0]        \n",
      "                                                                 conv5_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_0_bn (BatchNormali (None, 8, 8, 768)    3072        conv5_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_0_relu (Activation (None, 8, 8, 768)    0           conv5_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_conv (Conv2D)    (None, 8, 8, 128)    98304       conv5_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_bn (BatchNormali (None, 8, 8, 128)    512         conv5_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_relu (Activation (None, 8, 8, 128)    0           conv5_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_2_conv (Conv2D)    (None, 8, 8, 32)     36864       conv5_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_concat (Concatenat (None, 8, 8, 800)    0           conv5_block8_concat[0][0]        \n",
      "                                                                 conv5_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_0_bn (BatchNormal (None, 8, 8, 800)    3200        conv5_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_0_relu (Activatio (None, 8, 8, 800)    0           conv5_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_conv (Conv2D)   (None, 8, 8, 128)    102400      conv5_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_concat (Concatena (None, 8, 8, 832)    0           conv5_block9_concat[0][0]        \n",
      "                                                                 conv5_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_0_bn (BatchNormal (None, 8, 8, 832)    3328        conv5_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_0_relu (Activatio (None, 8, 8, 832)    0           conv5_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_conv (Conv2D)   (None, 8, 8, 128)    106496      conv5_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_concat (Concatena (None, 8, 8, 864)    0           conv5_block10_concat[0][0]       \n",
      "                                                                 conv5_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_0_bn (BatchNormal (None, 8, 8, 864)    3456        conv5_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_0_relu (Activatio (None, 8, 8, 864)    0           conv5_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_conv (Conv2D)   (None, 8, 8, 128)    110592      conv5_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_concat (Concatena (None, 8, 8, 896)    0           conv5_block11_concat[0][0]       \n",
      "                                                                 conv5_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_0_bn (BatchNormal (None, 8, 8, 896)    3584        conv5_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_0_relu (Activatio (None, 8, 8, 896)    0           conv5_block13_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_conv (Conv2D)   (None, 8, 8, 128)    114688      conv5_block13_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block13_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block13_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block13_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_concat (Concatena (None, 8, 8, 928)    0           conv5_block12_concat[0][0]       \n",
      "                                                                 conv5_block13_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_0_bn (BatchNormal (None, 8, 8, 928)    3712        conv5_block13_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_0_relu (Activatio (None, 8, 8, 928)    0           conv5_block14_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_conv (Conv2D)   (None, 8, 8, 128)    118784      conv5_block14_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block14_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block14_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block14_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_concat (Concatena (None, 8, 8, 960)    0           conv5_block13_concat[0][0]       \n",
      "                                                                 conv5_block14_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_0_bn (BatchNormal (None, 8, 8, 960)    3840        conv5_block14_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_0_relu (Activatio (None, 8, 8, 960)    0           conv5_block15_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_conv (Conv2D)   (None, 8, 8, 128)    122880      conv5_block15_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block15_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block15_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "conv5_block15_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block15_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_concat (Concatena (None, 8, 8, 992)    0           conv5_block14_concat[0][0]       \n",
      "                                                                 conv5_block15_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_0_bn (BatchNormal (None, 8, 8, 992)    3968        conv5_block15_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_0_relu (Activatio (None, 8, 8, 992)    0           conv5_block16_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_conv (Conv2D)   (None, 8, 8, 128)    126976      conv5_block16_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_bn (BatchNormal (None, 8, 8, 128)    512         conv5_block16_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_relu (Activatio (None, 8, 8, 128)    0           conv5_block16_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_2_conv (Conv2D)   (None, 8, 8, 32)     36864       conv5_block16_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_concat (Concatena (None, 8, 8, 1024)   0           conv5_block15_concat[0][0]       \n",
      "                                                                 conv5_block16_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "bn (BatchNormalization)         (None, 8, 8, 1024)   4096        conv5_block16_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "relu (Activation)               (None, 8, 8, 1024)   0           bn[0][0]                         \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d (AveragePooli (None, 2, 2, 1024)   0           relu[0][0]                       \n",
      "__________________________________________________________________________________________________\n",
      "flatten (Flatten)               (None, 4096)         0           average_pooling2d[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "dense (Dense)                   (None, 64)           262208      flatten[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dropout (Dropout)               (None, 64)           0           dense[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 3)            195         dropout[0][0]                    \n",
      "==================================================================================================\n",
      "Total params: 7,299,907\n",
      "Trainable params: 7,216,259\n",
      "Non-trainable params: 83,648\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# load the saved model\n",
    "saved_model = tf.keras.models.load_model('best_model.h5')\n",
    "saved_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "evaluate the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:  0.3835616409778595 , Test:  0.42105263471603394\n"
     ]
    }
   ],
   "source": [
    "#trainX, testX, trainY, testY\n",
    "# evaluate the model\n",
    "_,train_acc = saved_model.evaluate(trainX, trainY, verbose=0)\n",
    "_,test_acc = saved_model.evaluate(testX, testY, verbose=0)\n",
    "print('Train: ',train_acc,', Test: ' ,  test_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy', 'lr'])\n"
     ]
    }
   ],
   "source": [
    "history_dict = H.history\n",
    "print(history_dict.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "plot of training model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "H.history.keys()\n",
    "plt.plot(H.history['accuracy'])\n",
    "#plt.plot(H.history['val_accuracy'])\n",
    "plt.title('Accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train'], loc='upper left')\n",
    "\n",
    "plt.savefig( '{}_{}_epochs_{}_BS_{}_Accuracy_plot.png'.format(os.path.dirname(imagePaths[0]).split('/')[1],EPOCHS,BS,dt))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# shows epoches value\n",
    "len(H.history['val_accuracy']) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot confusion matrix for the TRAINING DATA results:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating the trained model ...\n",
      "                      precision    recall  f1-score   support\n",
      "\n",
      "    ./newtrain/covid       1.00      0.29      0.44         7\n",
      "   ./newtrain/normal       0.42      0.62      0.50         8\n",
      "./newtrain/pneumonia       0.20      0.25      0.22         4\n",
      "\n",
      "            accuracy                           0.42        19\n",
      "           macro avg       0.54      0.39      0.39        19\n",
      "        weighted avg       0.59      0.42      0.42        19\n",
      "\n",
      "total:  19\n",
      "confusion matrix: \n",
      " [[2 4 1]\n",
      " [0 5 3]\n",
      " [0 3 1]]\n",
      "acc: 0.4211\n",
      "sensitivity: 0.2857\n",
      "specificity: 1.0000\n"
     ]
    }
   ],
   "source": [
    "print(\"Evaluating the trained model ...\")\n",
    "predIdxs = saved_model.predict(testX, batch_size=BS)\n",
    "\n",
    "predIdxs = np.argmax(predIdxs, axis=1)\n",
    "\n",
    "print(classification_report(testY.argmax(axis=1), predIdxs, target_names=label_encoder.classes_))\n",
    "\n",
    "# calculate a basic confusion matrix\n",
    "cm = confusion_matrix(testY.argmax(axis=1), predIdxs)\n",
    "total = sum(sum(cm))\n",
    "print('total: ',total)\n",
    "acc = (cm[0, 0] + cm[1, 1] + cm[2, 2]) / total\n",
    "sensitivity = cm[0, 0] / (cm[0, 0] + cm[0, 1] + cm[0, 2])\n",
    "specificity = (cm[1, 1] + cm[1, 2] + cm[2, 1] + cm[2, 2]) / (cm[1, 0] + cm[1, 1] + cm[1, 2] + cm[2, 0] + cm[2, 1] + cm[2, 2])\n",
    "\n",
    "# show the confusion matrix, accuracy, sensitivity, and specificity\n",
    "print('confusion matrix: \\n',cm)\n",
    "print(\"acc: {:.4f}\".format(acc))\n",
    "print(\"sensitivity: {:.4f}\".format(sensitivity))\n",
    "print(\"specificity: {:.4f}\".format(specificity))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./newtrain/covid     .......covid ....0 ./newtrain/covid\n",
      "./newtrain/normal    ......normal ....1 ./newtrain/normal\n",
      "./newtrain/pneumonia ...pneumonia ....2 ./newtrain/pneumonia\n"
     ]
    }
   ],
   "source": [
    "#prediction_decoded = label_encoder.inverse_transform(predClasses)\n",
    "for i ,j, k, l in zip(list_of_classes,label_encoder.transform(list_of_classes),c2,label_encoder.classes_):\n",
    "    print('{:20}'.format(i),'{:.>12}'.format(k),'{:.>5}'.format(j),'{:.>5}'.format(l))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Test Dataset (17,958 CXR images) Confusion Matrix')]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion matrix using sklearn\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm,display_labels=c2)\n",
    "disp.plot()\n",
    "disp.ax_.set(\n",
    "                title='Test Dataset (17,958 CXR images) Confusion Matrix')#, "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion matrix using seaborn \n",
    "## The Seaborn Library provides data visualization. In this example, it plots\n",
    "## the Confusion Matrix\n",
    "import seaborn as sns\n",
    "## Call the heatmap() function from the Seaborn Library.\n",
    "## annot=True annotates cells.\n",
    "## fmt='g' disables scientific notation.\n",
    "## The heatmap() function returns a Matplotlib Axes Object.\n",
    "ax = sns.heatmap(cm, annot=True, fmt='g');\n",
    "## Modify the Axes Object directly to set various attributes such as the\n",
    "## Title, X/Y Labels.\n",
    "ax.set_title('Test Dataset (17,958 CXR images) Confusion Matrix');\n",
    "ax.set_xlabel('Predicted Label')\n",
    "ax.set_ylabel('Actual Label');\n",
    "## For the Tick Labels, the labels should be in Alphabetical order\n",
    "ax.xaxis.set_ticklabels(c2)\n",
    "ax.yaxis.set_ticklabels(c2)\n",
    "## Finally call the matplotlib show() function to display the visualization\n",
    "## of the Confusion Matrix.\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training and validation accuracy "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_training(history,EPOCHS,BS,dt):\n",
    "  acc = history.history['accuracy']\n",
    "  val_acc = history.history['val_accuracy']\n",
    "  loss = history.history['loss']\n",
    "  val_loss = history.history['val_loss']\n",
    "  epochs = range(len(acc))\n",
    "  \n",
    "  plt.plot(epochs, acc, 'r.')\n",
    "  plt.plot(epochs, val_acc, 'r')\n",
    "  plt.title('Training and validation accuracy')\n",
    "  #plt.savefig(\"./customDenseNet121_plot00a.png\")\n",
    "  plt.savefig( '{}_{}_epochs_{}_BS_{}_Training&validation accuracyplot.png'.format(os.path.dirname(imagePaths[0]).split('/')[1],EPOCHS,BS,dt))\n",
    "  plt.figure()\n",
    "  plt.plot(epochs, loss, 'r.')\n",
    "  plt.plot(epochs, val_loss, 'r-')\n",
    "  plt.title('Training and validation loss')\n",
    "  #plt.savefig(\"./customDenseNet121_plot00b.png\")\n",
    "  plt.savefig( '{}_{}_epochs_{}_BS_{}_Training&validationLossplot.png'.format(os.path.dirname(imagePaths[0]).split('/')[1],EPOCHS,BS,dt))\n",
    "  plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_training(H,EPOCHS,BS,dt)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot confusion matrix for testing some real X-rays "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# plot the training loss and accuracy\n",
    "N = EPOCHS \n",
    "plt.style.use(\"ggplot\")\n",
    "plt.figure()\n",
    "plt.plot(np.arange(0, N), H.history[\"loss\"], label=\"train_loss\",c ='aqua')\n",
    "plt.plot(np.arange(0, N), H.history[\"val_loss\"], label=\"val_loss\",c ='lime')\n",
    "plt.plot(np.arange(0, N), H.history[\"accuracy\"], label=\"train_accuracy\",c ='red')\n",
    "plt.plot(np.arange(0, N), H.history[\"val_accuracy\"], label=\"val_accuracy\",c ='b')\n",
    "plt.title(\"Training Loss and Accuracy on COVID-19 Training Dataset\")\n",
    "plt.xlabel(\"Epoch #\")\n",
    "plt.ylabel(\"Loss/Accuracy\")\n",
    "#plt.legend(bbox_to_anchor =(1.2, 1.25), ncol =4 )\n",
    "#plt.savefig(\"./customDenseNet121_plot01.png\")\n",
    "#plt.legend(loc=\"lower left\")\n",
    "plt.legend('loc=best')\n",
    "plt.savefig( '{}_{}_epochs_{}_BS_{}_plot.png'.format(os.path.dirname(imagePaths[0]).split('/')[1],EPOCHS,BS,dt))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Validation Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ellipsis"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "...\n",
    "imagePathValidation = \"./validation_2/\"\n",
    "#imagePathValidation = \"./validation/\"\n",
    "..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./validation_2/\n"
     ]
    }
   ],
   "source": [
    "print(imagePathValidation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] loading untrained test images...\n",
      "33\n"
     ]
    }
   ],
   "source": [
    "print(\"[INFO] loading untrained test images...\")\n",
    "#imagePathTest = \"./validation\"\n",
    "imagePathsValidation = list(paths.list_images(imagePathValidation))\n",
    "print(len(imagePathsValidation))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]]\n"
     ]
    }
   ],
   "source": [
    "dataValidation = []\n",
    "labelsValidation = []\n",
    "\n",
    "counter2 = 1\n",
    "filesize_v=[]\n",
    "# loop over the image paths\n",
    "for imagePathValidation in imagePathsValidation:\n",
    "    #print(counter2)\n",
    "    # extract the class label from the filename\n",
    "    labelValidation = imagePathValidation.split(os.path.sep)[-2]\n",
    "  \n",
    "\t# load the image, swap color channels, and resize it to be a fixed\n",
    "\t# 256x256 pixels while ignoring aspect ratio\n",
    "    imageValidation = cv2.imread(imagePathValidation)\n",
    "    imageValidation = cv2.cvtColor(imageValidation, cv2.COLOR_BGR2RGB)\n",
    "    height_v, width_v, channels_v = imageValidation.shape\n",
    "    filesize_v.append([labelValidation,height_v,width_v,channels_v,height_v*width_v*channels_v])\n",
    "    imageValidation = cv2.resize(imageValidation, (imagesize, imagesize))\n",
    "\n",
    "\t# update the data and labels lists, respectively\n",
    "    dataValidation.append(imageValidation)\n",
    "    labelsValidation.append(labelValidation)\n",
    "    counter2=counter2+1\n",
    "    \n",
    "# convert the data and labels to NumPy arrays while scaling the pixel\n",
    "# intensities to the range [0, 255]\n",
    "dataValidation = np.array(dataValidation) / 255.0\n",
    "labelsValidation = np.array(labelsValidation)\n",
    "\n",
    "\n",
    "# perform one-hot encoding on the labels\n",
    "#from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "#label_encoder2 = LabelEncoder()\n",
    "integer_encoded2 = label_encoder.fit_transform(labelsValidation)\n",
    "labelsValidation = to_categorical(integer_encoded2)\n",
    "\n",
    "print(labelsValidation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label_v</th>\n",
       "      <th>height_v</th>\n",
       "      <th>width_v</th>\n",
       "      <th>channels_v</th>\n",
       "      <th>size_v</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>./validation_2/covid</td>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>./validation_2/normal</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>./validation_2/pneumonia</td>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     label_v  height_v  width_v  channels_v   size_v\n",
       "0       ./validation_2/covid       512      512           3   786432\n",
       "1       ./validation_2/covid       512      512           3   786432\n",
       "2       ./validation_2/covid       512      512           3   786432\n",
       "3       ./validation_2/covid       512      512           3   786432\n",
       "4       ./validation_2/covid       512      512           3   786432\n",
       "5       ./validation_2/covid       512      512           3   786432\n",
       "6       ./validation_2/covid       512      512           3   786432\n",
       "7       ./validation_2/covid       512      512           3   786432\n",
       "8       ./validation_2/covid       512      512           3   786432\n",
       "9       ./validation_2/covid       512      512           3   786432\n",
       "10      ./validation_2/covid       512      512           3   786432\n",
       "11     ./validation_2/normal      1024     1024           3  3145728\n",
       "12     ./validation_2/normal      1024     1024           3  3145728\n",
       "13     ./validation_2/normal      1024     1024           3  3145728\n",
       "14     ./validation_2/normal      1024     1024           3  3145728\n",
       "15     ./validation_2/normal      1024     1024           3  3145728\n",
       "16     ./validation_2/normal      1024     1024           3  3145728\n",
       "17     ./validation_2/normal      1024     1024           3  3145728\n",
       "18     ./validation_2/normal      1024     1024           3  3145728\n",
       "19     ./validation_2/normal      1024     1024           3  3145728\n",
       "20     ./validation_2/normal      1024     1024           3  3145728\n",
       "21     ./validation_2/normal      1024     1024           3  3145728\n",
       "22  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "23  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "24  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "25  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "26  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "27  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "28  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "29  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "30  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "31  ./validation_2/pneumonia      1024     1024           3  3145728\n",
       "32  ./validation_2/pneumonia      1024     1024           3  3145728"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_v=pd.DataFrame(filesize_v,columns=[\"label_v\",\"height_v\",\"width_v\",\"channels_v\",\"size_v\"])\n",
    "df_v"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.DataFrame(df_v.value_counts()).to_csv('validation_image_size.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['label_v', 'height_v', 'width_v', 'channels_v', 'size_v', '0']\n",
      "['./validation_2/covid', '512', '512', '3', '786432', '11']\n",
      "['./validation_2/normal', '1024', '1024', '3', '3145728', '11']\n",
      "['./validation_2/pneumonia', '1024', '1024', '3', '3145728', '11']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "with open('validation_image_size.csv', newline='') as f:\n",
    "    reader = csv.reader(f)\n",
    "    for row in reader:\n",
    "        print(row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " size of data:  6488064 \n",
      " No of rows:  33 \n",
      " No of labels:  33 \n",
      " Type:  <class 'numpy.ndarray'> \n",
      " rows:  33 \n",
      " columns:  256 \n",
      " total_length:  8448\n"
     ]
    }
   ],
   "source": [
    "print(' size of data: ',dataValidation.size, '\\n No of rows: ',len(dataValidation),'\\n No of labels: ',len(labelsValidation),'\\n Type: ',type(dataValidation),'\\n','rows: ',len(dataValidation),'\\n','columns: ',len(dataValidation[0]),'\\n','total_length: ',len(dataValidation)*len(dataValidation[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "33 256 256 3\n"
     ]
    }
   ],
   "source": [
    "print(len(dataValidation),len(dataValidation[0]),len(dataValidation[0][0]),len(dataValidation[0][0][0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.13485663 0.4202453  0.4448981 ]\n",
      " [0.06006541 0.60165536 0.33827925]\n",
      " [0.11307012 0.3140603  0.57286954]\n",
      " [0.3670434  0.2691913  0.36376536]\n",
      " [0.04234363 0.5297979  0.42785847]\n",
      " [0.00931759 0.842054   0.14862835]\n",
      " [0.04699151 0.39835137 0.55465704]\n",
      " [0.03544186 0.47016218 0.49439594]\n",
      " [0.03773639 0.23222497 0.73003864]\n",
      " [0.10953868 0.50858706 0.38187423]\n",
      " [0.02096627 0.65081364 0.3282201 ]\n",
      " [0.0371626  0.647243   0.31559438]\n",
      " [0.03421749 0.65866286 0.30711967]\n",
      " [0.06676361 0.0992268  0.8340095 ]\n",
      " [0.08175757 0.18607497 0.7321675 ]\n",
      " [0.11608361 0.61376363 0.27015275]\n",
      " [0.16550806 0.36823606 0.46625587]\n",
      " [0.08241363 0.3795789  0.5380075 ]\n",
      " [0.04749401 0.650593   0.30191302]\n",
      " [0.08723078 0.39218163 0.52058756]\n",
      " [0.0765122  0.6204647  0.3030231 ]\n",
      " [0.03171171 0.60440356 0.36388466]\n",
      " [0.12043834 0.5644603  0.3151014 ]\n",
      " [0.19744162 0.25545344 0.5471049 ]\n",
      " [0.22891411 0.4527491  0.31833673]\n",
      " [0.36749002 0.37136462 0.26114535]\n",
      " [0.14531422 0.6966043  0.15808152]\n",
      " [0.06771268 0.6692339  0.26305342]\n",
      " [0.22212614 0.2667738  0.5111    ]\n",
      " [0.08407763 0.7497078  0.16621453]\n",
      " [0.38144535 0.42548722 0.19306743]\n",
      " [0.07950857 0.6904296  0.23006183]\n",
      " [0.10713368 0.5534399  0.3394265 ]]\n",
      "[2 1 2 0 1 1 2 2 2 1 1 1 1 2 2 1 2 2 1 2 1 1 1 2 1 1 1 1 2 1 1 1 1]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Ellipsis"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predValidation = saved_model.predict(dataValidation, batch_size=BS)\n",
    "\n",
    "print(predValidation)\n",
    "\n",
    "predValidationClasses = predValidation.argmax(axis=-1)\n",
    "print(predValidationClasses)\n",
    "..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluating the trained model ...\n",
      "                          precision    recall  f1-score   support\n",
      "\n",
      "    ./validation_2/covid       1.00      0.09      0.17        11\n",
      "   ./validation_2/normal       0.30      0.55      0.39        11\n",
      "./validation_2/pneumonia       0.17      0.18      0.17        11\n",
      "\n",
      "                accuracy                           0.27        33\n",
      "               macro avg       0.49      0.27      0.24        33\n",
      "            weighted avg       0.49      0.27      0.24        33\n",
      "\n",
      "[[1 5 5]\n",
      " [0 6 5]\n",
      " [0 9 2]]\n",
      "acc: 0.2727\n",
      "sensitivity: 0.0909\n",
      "specificity: 1.0000\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "testX_v = dataValidation\n",
    "testY_v = labelsValidation\n",
    "\n",
    "print(\"Evaluating the trained model ...\")\n",
    "predIdxs_v = saved_model.predict(testX_v, batch_size=BS)\n",
    "\n",
    "predIdxs_v = np.argmax(predIdxs_v, axis=1)\n",
    "\n",
    "print(classification_report(testY_v.argmax(axis=1), predIdxs_v, target_names=label_encoder.classes_))\n",
    "\n",
    "# calculate a basic confusion matrix\n",
    "cm2 = confusion_matrix(testY_v.argmax(axis=1), predIdxs_v)\n",
    "total2 = sum(sum(cm2))\n",
    "acc2 = (cm2[0, 0] + cm2[1, 1] + cm2[2, 2]) / total2\n",
    "sensitivity2 = cm2[0, 0] / (cm2[0, 0] + cm2[0, 1] + cm2[0, 2])\n",
    "specificity2 = (cm2[1, 1] + cm2[1, 2] + cm2[2, 1] + cm2[2, 2]) / (cm2[1, 0] + cm2[1, 1] + cm2[1, 2] + cm2[2, 0] + cm2[2, 1] + cm2[2, 2])\n",
    "\n",
    "# show the confusion matrix, accuracy, sensitivity, and specificity\n",
    "print(cm2)\n",
    "print(\"acc: {:.4f}\".format(acc2))\n",
    "print(\"sensitivity: {:.4f}\".format(sensitivity2))\n",
    "print(\"specificity: {:.4f}\".format(specificity2))\n",
    "\n",
    "# plot the training loss and accuracy\n",
    "N = EPOCHS \n",
    "plt.style.use(\"ggplot\")\n",
    "plt.figure()\n",
    "\n",
    "plt.plot(np.arange(0, N), H.history[\"loss\"], label=\"train_loss\",c ='aqua')\n",
    "plt.plot(np.arange(0, N), H.history[\"val_loss\"], label=\"val_loss\",c ='lime')\n",
    "plt.plot(np.arange(0, N), H.history[\"accuracy\"], label=\"train_accuracy\",c ='red')\n",
    "plt.plot(np.arange(0, N), H.history[\"val_accuracy\"], label=\"val_accuracy\",c ='b')\n",
    "\n",
    "#plt.plot(np.arange(0, N), H.history[\"loss\"], label=\"train_loss\")\n",
    "#plt.plot(np.arange(0, N), H.history[\"val_loss\"], label=\"val_loss\")\n",
    "#plt.plot(np.arange(0, N), H.history[\"acc\"], label=\"train_acc\")\n",
    "#plt.plot(np.arange(0, N), H.history[\"val_acc\"], label=\"val_acc\")\n",
    "plt.title(\"Training Loss and Accuracy on COVID-19 Validation Dataset\")\n",
    "plt.xlabel(\"Epoch #\")\n",
    "plt.ylabel(\"Loss/Accuracy\")\n",
    "plt.legend(loc='best')\n",
    "#plt.savefig(\"./Validation_Data.png\")\n",
    "plt.savefig( '{}_{}_epochs_{}_BS_{}plot.png'.format(os.path.dirname(imagePathsValidation[0]).split('/')[1],EPOCHS,BS,dt))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Validation Dataset (3,432 CXR images) Confusion Matrix')]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion matrix of validation dataset from sklearn\n",
    "disp3 = ConfusionMatrixDisplay(confusion_matrix=cm2,display_labels=c2)\n",
    "disp3.plot()\n",
    "disp3.ax_.set(\n",
    "                title='Validation Dataset (3,432 CXR images) Confusion Matrix')#, \n",
    "                #xlabel='Predicted Fruits', \n",
    "                #ylabel='Actual Fruits')\n",
    "#disp.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion matrix of validation dataset from seaborn\n",
    "## The Seaborn Library provides data visualization. In this example, it plots\n",
    "## the Confusion Matrix\n",
    "import seaborn as sns\n",
    "## Call the heatmap() function from the Seaborn Library.\n",
    "## annot=True annotates cells.\n",
    "## fmt='g' disables scientific notation.\n",
    "## The heatmap() function returns a Matplotlib Axes Object.\n",
    "ax = sns.heatmap(cm2, annot=True, fmt='g');\n",
    "## Modify the Axes Object directly to set various attributes such as the\n",
    "## Title, X/Y Labels.\n",
    "ax.set_title('Validation Dataset (3,432 CXR images) Confusion Matrix');\n",
    "ax.set_xlabel('Predicted Label')\n",
    "ax.set_ylabel('Actual Label');\n",
    "## For the Tick Labels, the labels should be in Alphabetical order\n",
    "ax.xaxis.set_ticklabels(c2)\n",
    "ax.yaxis.set_ticklabels(c2)\n",
    "## Finally call the matplotlib show() function to display the visualization\n",
    "## of the Confusion Matrix.\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ellipsis"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "...\n",
    "imagePathTest = \"./test_set_2/\"\n",
    "..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./test_set_2/\n"
     ]
    }
   ],
   "source": [
    "print(imagePathTest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] loading untrained test images...\n",
      "52\n"
     ]
    }
   ],
   "source": [
    "print(\"[INFO] loading untrained test images...\")\n",
    "#imagePathTest = \"./validation\"\n",
    "imagePathsTest = list(paths.list_images(imagePathTest))\n",
    "print(len(imagePathsTest))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataTest = []\n",
    "labelsTest = []\n",
    "\n",
    "counter2 = 1\n",
    "filesize_test=[]\n",
    "# loop over the image paths\n",
    "for imagePathTest in imagePathsTest:\n",
    "    #print(counter2)\n",
    "    # extract the class label from the filename\n",
    "  #  labelTest = imagePathTest.split(os.path.sep)[-2]\n",
    "  \n",
    "\t# load the image, swap color channels, and resize it to be a fixed\n",
    "\t# 256x256 pixels while ignoring aspect ratio\n",
    "    imageTest = cv2.imread(imagePathTest)\n",
    "    imageTest = cv2.cvtColor(imageTest, cv2.COLOR_BGR2RGB)\n",
    "    height_test, width_test, channels_test = imageTest.shape\n",
    "    filesize_test.append([height_test,width_test,channels_test,height_test*width_test*channels_test])\n",
    "    imageTest = cv2.resize(imageTest, (imagesize, imagesize))\n",
    "\n",
    "\t# update the data and labels lists, respectively\n",
    "    dataTest.append(imageTest)\n",
    "   # labelsTest.append(labelTest)\n",
    "    counter2=counter2+1\n",
    "    \n",
    "# convert the data and labels to NumPy arrays while scaling the pixel\n",
    "# intensities to the range [0, 255]\n",
    "dataTest = np.array(dataTest) / 255.0\n",
    "#labelsTest = np.array(labelsTest)\n",
    "\n",
    "\n",
    "# perform one-hot encoding on the labels\n",
    "#from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "#label_encoder2 = LabelEncoder()\n",
    "#integer_encoded2 = label_encoder2.fit_transform(labelsTest)\n",
    "#labelsTest = to_categorical(integer_encoded2)\n",
    "\n",
    "#print(labelsTest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>height_test</th>\n",
       "      <th>width_test</th>\n",
       "      <th>channels_test</th>\n",
       "      <th>size_test</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>512</td>\n",
       "      <td>512</td>\n",
       "      <td>3</td>\n",
       "      <td>786432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>1024</td>\n",
       "      <td>1024</td>\n",
       "      <td>3</td>\n",
       "      <td>3145728</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    height_test  width_test  channels_test  size_test\n",
       "0          1024        1024              3    3145728\n",
       "1          1024        1024              3    3145728\n",
       "2          1024        1024              3    3145728\n",
       "3          1024        1024              3    3145728\n",
       "4           512         512              3     786432\n",
       "5          1024        1024              3    3145728\n",
       "6           512         512              3     786432\n",
       "7          1024        1024              3    3145728\n",
       "8           512         512              3     786432\n",
       "9          1024        1024              3    3145728\n",
       "10          512         512              3     786432\n",
       "11         1024        1024              3    3145728\n",
       "12         1024        1024              3    3145728\n",
       "13          512         512              3     786432\n",
       "14          512         512              3     786432\n",
       "15         1024        1024              3    3145728\n",
       "16          512         512              3     786432\n",
       "17         1024        1024              3    3145728\n",
       "18         1024        1024              3    3145728\n",
       "19         1024        1024              3    3145728\n",
       "20         1024        1024              3    3145728\n",
       "21          512         512              3     786432\n",
       "22          512         512              3     786432\n",
       "23         1024        1024              3    3145728\n",
       "24          512         512              3     786432\n",
       "25         1024        1024              3    3145728\n",
       "26         1024        1024              3    3145728\n",
       "27         1024        1024              3    3145728\n",
       "28          512         512              3     786432\n",
       "29         1024        1024              3    3145728\n",
       "30         1024        1024              3    3145728\n",
       "31         1024        1024              3    3145728\n",
       "32         1024        1024              3    3145728\n",
       "33         1024        1024              3    3145728\n",
       "34         1024        1024              3    3145728\n",
       "35          512         512              3     786432\n",
       "36          512         512              3     786432\n",
       "37         1024        1024              3    3145728\n",
       "38         1024        1024              3    3145728\n",
       "39         1024        1024              3    3145728\n",
       "40          512         512              3     786432\n",
       "41          512         512              3     786432\n",
       "42          512         512              3     786432\n",
       "43         1024        1024              3    3145728\n",
       "44          512         512              3     786432\n",
       "45         1024        1024              3    3145728\n",
       "46         1024        1024              3    3145728\n",
       "47          512         512              3     786432\n",
       "48         1024        1024              3    3145728\n",
       "49          512         512              3     786432\n",
       "50         1024        1024              3    3145728\n",
       "51         1024        1024              3    3145728"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test=pd.DataFrame(filesize_test,columns=[\"height_test\",\"width_test\",\"channels_test\",\"size_test\"])\n",
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.DataFrame(df_test.value_counts()).to_csv('test_image_size.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['height_test', 'width_test', 'channels_test', 'size_test', '0']\n",
      "['1024', '1024', '3', '3145728', '33']\n",
      "['512', '512', '3', '786432', '19']\n"
     ]
    }
   ],
   "source": [
    "with open('test_image_size.csv', newline='') as f:\n",
    "    reader = csv.reader(f)\n",
    "    for row in reader:\n",
    "        print(row)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " size of data:  10223616 \n",
      " No of rows:  52 \n",
      " Type:  <class 'numpy.ndarray'> \n",
      " rows:  52 \n",
      " columns:  256 \n",
      " total_length:  13312\n"
     ]
    }
   ],
   "source": [
    "print(' size of data: ',dataTest.size, '\\n No of rows: ',len(dataTest),\n",
    "      #'\\n No of labels: ',len(labelsTest),\n",
    "      '\\n Type: ',type(dataTest),'\\n','rows: ',len(dataTest),'\\n','columns: ',len(dataTest[0]),'\\n','total_length: ',len(dataTest)*len(dataTest[0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "52 256 256 3\n"
     ]
    }
   ],
   "source": [
    "print(len(dataTest),len(dataTest[0]),len(dataTest[0][0]),len(dataTest[0][0][0]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.02800864 0.65353763 0.31845376]\n",
      " [0.3287969  0.26768014 0.403523  ]\n",
      " [0.05309611 0.42475903 0.5221449 ]\n",
      " [0.22144128 0.4353643  0.34319443]\n",
      " [0.06521858 0.63210756 0.30267385]\n",
      " [0.0526653  0.63386416 0.31347048]\n",
      " [0.12898587 0.28797764 0.5830365 ]\n",
      " [0.35664156 0.30784923 0.33550924]\n",
      " [0.5586719  0.19304892 0.24827918]\n",
      " [0.21915211 0.48083594 0.30001193]\n",
      " [0.15743805 0.4947579  0.3478041 ]\n",
      " [0.10504591 0.4813342  0.41361982]\n",
      " [0.13610679 0.55752623 0.30636698]\n",
      " [0.04786373 0.69605035 0.2560859 ]\n",
      " [0.05811507 0.56891793 0.37296703]\n",
      " [0.1871753  0.58985585 0.22296885]\n",
      " [0.13421261 0.4631278  0.40265954]\n",
      " [0.03970136 0.64672554 0.31357306]\n",
      " [0.2981435  0.36536866 0.33648783]\n",
      " [0.49423045 0.31147328 0.19429621]\n",
      " [0.20866702 0.36772162 0.4236113 ]\n",
      " [0.05186068 0.5345474  0.41359198]\n",
      " [0.02656841 0.62421316 0.3492185 ]\n",
      " [0.1068607  0.45839158 0.43474767]\n",
      " [0.29094496 0.29103756 0.41801745]\n",
      " [0.21542615 0.50010365 0.28447017]\n",
      " [0.18597986 0.5300851  0.2839351 ]\n",
      " [0.10524614 0.43393788 0.46081597]\n",
      " [0.06867611 0.51766753 0.41365635]\n",
      " [0.2789154  0.42256832 0.2985164 ]\n",
      " [0.10567164 0.47397423 0.4203542 ]\n",
      " [0.08570357 0.68537325 0.22892316]\n",
      " [0.17416734 0.38762188 0.4382107 ]\n",
      " [0.08580124 0.556519   0.35767975]\n",
      " [0.12624066 0.63468087 0.23907846]\n",
      " [0.47573912 0.23104745 0.29321343]\n",
      " [0.06256528 0.4451003  0.49233446]\n",
      " [0.191957   0.47698277 0.33106032]\n",
      " [0.13828139 0.6255635  0.23615511]\n",
      " [0.03289966 0.17748067 0.7896196 ]\n",
      " [0.10724661 0.51287585 0.37987748]\n",
      " [0.08472684 0.32545355 0.5898196 ]\n",
      " [0.1346634  0.6057542  0.25958234]\n",
      " [0.02988583 0.668355   0.30175918]\n",
      " [0.07402314 0.42756855 0.4984083 ]\n",
      " [0.1180947  0.48737139 0.39453393]\n",
      " [0.02747975 0.49296728 0.479553  ]\n",
      " [0.04938577 0.3963776  0.55423665]\n",
      " [0.07242654 0.62178725 0.30578616]\n",
      " [0.03140518 0.55173063 0.4168642 ]\n",
      " [0.13266246 0.504959   0.3623785 ]\n",
      " [0.2030561  0.47561792 0.32132602]]\n",
      "[1 2 2 1 1 1 2 0 0 1 1 1 1 1 1 1 1 1 1 0 2 1 1 1 2 1 1 2 1 1 1 1 2 1 1 0 2\n",
      " 1 1 2 1 2 1 1 2 1 1 2 1 1 1 1]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Ellipsis"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predTest = saved_model.predict(dataTest, batch_size=BS)\n",
    "\n",
    "print(predTest)\n",
    "\n",
    "predClasses = predTest.argmax(axis=-1)\n",
    "print(predClasses)\n",
    "..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   0\n",
      "0  1\n",
      "1  2\n",
      "2  2\n",
      "3  1\n",
      "4  1\n",
      "(52,)\n",
      "52\n",
      "<class 'numpy.ndarray'>\n",
      "\n",
      "1    36\n",
      "2    12\n",
      "0     4\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "import pandas\n",
    "print(pd.DataFrame(predClasses).head())\n",
    "print(predClasses.shape)\n",
    "print(len(predClasses))\n",
    "print(type(predClasses))\n",
    "print()\n",
    "print(pd.DataFrame(predClasses).value_counts())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  Final File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "./test_set_2/0001.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0002.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0003.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0004.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0005.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0006.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0007.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0008.jpg  : 0 :  ./validation_2/covid\n",
      "./test_set_2/0009.jpg  : 0 :  ./validation_2/covid\n",
      "./test_set_2/0010.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0011.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0012.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0013.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0014.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0015.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0016.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0017.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0018.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0019.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0020.jpg  : 0 :  ./validation_2/covid\n",
      "./test_set_2/0021.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0022.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0023.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0024.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0025.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0026.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0027.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0028.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0029.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0030.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0031.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0032.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0033.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0034.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0035.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0036.jpg  : 0 :  ./validation_2/covid\n",
      "./test_set_2/0037.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0038.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0039.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0040.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0041.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0042.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0043.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0044.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0045.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0046.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0047.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0048.jpg  : 2 :  ./validation_2/pneumonia\n",
      "./test_set_2/0049.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0050.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0051.jpg  : 1 :  ./validation_2/normal\n",
      "./test_set_2/0052.jpg  : 1 :  ./validation_2/normal\n"
     ]
    }
   ],
   "source": [
    "import csv\n",
    "import re\n",
    "# name of csv file \n",
    "fn=\"submission2_\"\n",
    "filename = fn+dt+\".csv\"\n",
    "# open the file in the write mode\n",
    "f = open('/csv_file', 'w')\n",
    "prediction_decoded = label_encoder.inverse_transform(predClasses)\n",
    "\n",
    "#prediction_decoded2 = labels.inverse_transform(predClasses)\n",
    "#print(prediction_decoded)\n",
    "# writing to csv file \n",
    "with open(filename, 'w', newline='') as csvfile: \n",
    "    # creating a csv writer object \n",
    "    csvwriter = csv.writer(csvfile) \n",
    "    csvwriter.writerow(['case', 'class'])     \n",
    "    for i,j,k in zip(imagePathsTest,predClasses,prediction_decoded):\n",
    "        print(re.sub(r'^./test_set/','',i),' :',j,': ',re.sub(r'^./newtrain/','',k))\n",
    "        # writing the fields \n",
    "        #csvwriter.writerow((re.sub(r'^./test_set/','',i),re.sub(r'^./newtrain/','',k))) \n",
    "        csvwriter.writerow((re.sub(r'^./test_set/','',i),j)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>case</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./test_set_2/0001.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./test_set_2/0002.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./test_set_2/0003.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./test_set_2/0004.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./test_set_2/0005.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>./test_set_2/0006.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>./test_set_2/0007.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>./test_set_2/0008.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>./test_set_2/0009.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>./test_set_2/0010.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>./test_set_2/0011.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>./test_set_2/0012.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>./test_set_2/0013.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>./test_set_2/0014.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>./test_set_2/0015.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>./test_set_2/0016.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>./test_set_2/0017.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>./test_set_2/0018.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>./test_set_2/0019.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>./test_set_2/0020.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>./test_set_2/0021.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>./test_set_2/0022.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>./test_set_2/0023.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>./test_set_2/0024.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>./test_set_2/0025.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>./test_set_2/0026.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>./test_set_2/0027.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>./test_set_2/0028.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>./test_set_2/0029.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>./test_set_2/0030.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>./test_set_2/0031.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>./test_set_2/0032.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>./test_set_2/0033.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>./test_set_2/0034.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>./test_set_2/0035.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>./test_set_2/0036.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>./test_set_2/0037.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>./test_set_2/0038.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>./test_set_2/0039.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>./test_set_2/0040.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>./test_set_2/0041.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>./test_set_2/0042.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>./test_set_2/0043.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>./test_set_2/0044.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>./test_set_2/0045.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>./test_set_2/0046.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>./test_set_2/0047.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>./test_set_2/0048.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>./test_set_2/0049.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>./test_set_2/0050.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>./test_set_2/0051.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>./test_set_2/0052.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     case  class\n",
       "0   ./test_set_2/0001.jpg      1\n",
       "1   ./test_set_2/0002.jpg      2\n",
       "2   ./test_set_2/0003.jpg      2\n",
       "3   ./test_set_2/0004.jpg      1\n",
       "4   ./test_set_2/0005.jpg      1\n",
       "5   ./test_set_2/0006.jpg      1\n",
       "6   ./test_set_2/0007.jpg      2\n",
       "7   ./test_set_2/0008.jpg      0\n",
       "8   ./test_set_2/0009.jpg      0\n",
       "9   ./test_set_2/0010.jpg      1\n",
       "10  ./test_set_2/0011.jpg      1\n",
       "11  ./test_set_2/0012.jpg      1\n",
       "12  ./test_set_2/0013.jpg      1\n",
       "13  ./test_set_2/0014.jpg      1\n",
       "14  ./test_set_2/0015.jpg      1\n",
       "15  ./test_set_2/0016.jpg      1\n",
       "16  ./test_set_2/0017.jpg      1\n",
       "17  ./test_set_2/0018.jpg      1\n",
       "18  ./test_set_2/0019.jpg      1\n",
       "19  ./test_set_2/0020.jpg      0\n",
       "20  ./test_set_2/0021.jpg      2\n",
       "21  ./test_set_2/0022.jpg      1\n",
       "22  ./test_set_2/0023.jpg      1\n",
       "23  ./test_set_2/0024.jpg      1\n",
       "24  ./test_set_2/0025.jpg      2\n",
       "25  ./test_set_2/0026.jpg      1\n",
       "26  ./test_set_2/0027.jpg      1\n",
       "27  ./test_set_2/0028.jpg      2\n",
       "28  ./test_set_2/0029.jpg      1\n",
       "29  ./test_set_2/0030.jpg      1\n",
       "30  ./test_set_2/0031.jpg      1\n",
       "31  ./test_set_2/0032.jpg      1\n",
       "32  ./test_set_2/0033.jpg      2\n",
       "33  ./test_set_2/0034.jpg      1\n",
       "34  ./test_set_2/0035.jpg      1\n",
       "35  ./test_set_2/0036.jpg      0\n",
       "36  ./test_set_2/0037.jpg      2\n",
       "37  ./test_set_2/0038.jpg      1\n",
       "38  ./test_set_2/0039.jpg      1\n",
       "39  ./test_set_2/0040.jpg      2\n",
       "40  ./test_set_2/0041.jpg      1\n",
       "41  ./test_set_2/0042.jpg      2\n",
       "42  ./test_set_2/0043.jpg      1\n",
       "43  ./test_set_2/0044.jpg      1\n",
       "44  ./test_set_2/0045.jpg      2\n",
       "45  ./test_set_2/0046.jpg      1\n",
       "46  ./test_set_2/0047.jpg      1\n",
       "47  ./test_set_2/0048.jpg      2\n",
       "48  ./test_set_2/0049.jpg      1\n",
       "49  ./test_set_2/0050.jpg      1\n",
       "50  ./test_set_2/0051.jpg      1\n",
       "51  ./test_set_2/0052.jpg      1"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#df = pandas.read_csv('submission.csv')\n",
    "df = pd.read_csv(filename)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                     case  class\n",
      "0   ./test_set_2/0001.jpg      1\n",
      "1   ./test_set_2/0002.jpg      2\n",
      "2   ./test_set_2/0003.jpg      2\n",
      "3   ./test_set_2/0004.jpg      1\n",
      "4   ./test_set_2/0005.jpg      1\n",
      "5   ./test_set_2/0006.jpg      1\n",
      "6   ./test_set_2/0007.jpg      2\n",
      "7   ./test_set_2/0008.jpg      0\n",
      "8   ./test_set_2/0009.jpg      0\n",
      "9   ./test_set_2/0010.jpg      1\n",
      "10  ./test_set_2/0011.jpg      1\n",
      "11  ./test_set_2/0012.jpg      1\n",
      "12  ./test_set_2/0013.jpg      1\n",
      "13  ./test_set_2/0014.jpg      1\n",
      "14  ./test_set_2/0015.jpg      1\n",
      "15  ./test_set_2/0016.jpg      1\n",
      "16  ./test_set_2/0017.jpg      1\n",
      "17  ./test_set_2/0018.jpg      1\n",
      "18  ./test_set_2/0019.jpg      1\n",
      "19  ./test_set_2/0020.jpg      0\n",
      "20  ./test_set_2/0021.jpg      2\n",
      "21  ./test_set_2/0022.jpg      1\n",
      "22  ./test_set_2/0023.jpg      1\n",
      "23  ./test_set_2/0024.jpg      1\n",
      "24  ./test_set_2/0025.jpg      2\n",
      "25  ./test_set_2/0026.jpg      1\n",
      "26  ./test_set_2/0027.jpg      1\n",
      "27  ./test_set_2/0028.jpg      2\n",
      "28  ./test_set_2/0029.jpg      1\n",
      "29  ./test_set_2/0030.jpg      1\n",
      "30  ./test_set_2/0031.jpg      1\n",
      "31  ./test_set_2/0032.jpg      1\n",
      "32  ./test_set_2/0033.jpg      2\n",
      "33  ./test_set_2/0034.jpg      1\n",
      "34  ./test_set_2/0035.jpg      1\n",
      "35  ./test_set_2/0036.jpg      0\n",
      "36  ./test_set_2/0037.jpg      2\n",
      "37  ./test_set_2/0038.jpg      1\n",
      "38  ./test_set_2/0039.jpg      1\n",
      "39  ./test_set_2/0040.jpg      2\n",
      "40  ./test_set_2/0041.jpg      1\n",
      "41  ./test_set_2/0042.jpg      2\n",
      "42  ./test_set_2/0043.jpg      1\n",
      "43  ./test_set_2/0044.jpg      1\n",
      "44  ./test_set_2/0045.jpg      2\n",
      "45  ./test_set_2/0046.jpg      1\n",
      "46  ./test_set_2/0047.jpg      1\n",
      "47  ./test_set_2/0048.jpg      2\n",
      "48  ./test_set_2/0049.jpg      1\n",
      "49  ./test_set_2/0050.jpg      1\n",
      "50  ./test_set_2/0051.jpg      1\n",
      "51  ./test_set_2/0052.jpg      1\n"
     ]
    }
   ],
   "source": [
    "file2=fn+dt+\".csv\"\n",
    "df = pd.read_csv(file2)\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>case</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>./test_set_2/0001.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>./test_set_2/0002.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>./test_set_2/0003.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>./test_set_2/0004.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>./test_set_2/0005.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>./test_set_2/0006.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>./test_set_2/0007.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>./test_set_2/0008.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>./test_set_2/0009.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>./test_set_2/0010.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>./test_set_2/0011.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>./test_set_2/0012.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>./test_set_2/0013.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>./test_set_2/0014.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>./test_set_2/0015.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>./test_set_2/0016.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>./test_set_2/0017.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>./test_set_2/0018.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>./test_set_2/0019.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>./test_set_2/0020.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>./test_set_2/0021.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>./test_set_2/0022.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>./test_set_2/0023.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>./test_set_2/0024.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>./test_set_2/0025.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>./test_set_2/0026.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>./test_set_2/0027.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>./test_set_2/0028.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>./test_set_2/0029.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>./test_set_2/0030.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>./test_set_2/0031.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>./test_set_2/0032.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>./test_set_2/0033.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>./test_set_2/0034.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>./test_set_2/0035.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>./test_set_2/0036.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>./test_set_2/0037.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>./test_set_2/0038.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>./test_set_2/0039.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>./test_set_2/0040.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>./test_set_2/0041.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>./test_set_2/0042.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>./test_set_2/0043.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>./test_set_2/0044.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>./test_set_2/0045.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>./test_set_2/0046.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>./test_set_2/0047.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>./test_set_2/0048.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>./test_set_2/0049.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>./test_set_2/0050.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>./test_set_2/0051.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>./test_set_2/0052.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     case  class\n",
       "0   ./test_set_2/0001.jpg      1\n",
       "1   ./test_set_2/0002.jpg      2\n",
       "2   ./test_set_2/0003.jpg      2\n",
       "3   ./test_set_2/0004.jpg      1\n",
       "4   ./test_set_2/0005.jpg      1\n",
       "5   ./test_set_2/0006.jpg      1\n",
       "6   ./test_set_2/0007.jpg      2\n",
       "7   ./test_set_2/0008.jpg      0\n",
       "8   ./test_set_2/0009.jpg      0\n",
       "9   ./test_set_2/0010.jpg      1\n",
       "10  ./test_set_2/0011.jpg      1\n",
       "11  ./test_set_2/0012.jpg      1\n",
       "12  ./test_set_2/0013.jpg      1\n",
       "13  ./test_set_2/0014.jpg      1\n",
       "14  ./test_set_2/0015.jpg      1\n",
       "15  ./test_set_2/0016.jpg      1\n",
       "16  ./test_set_2/0017.jpg      1\n",
       "17  ./test_set_2/0018.jpg      1\n",
       "18  ./test_set_2/0019.jpg      1\n",
       "19  ./test_set_2/0020.jpg      0\n",
       "20  ./test_set_2/0021.jpg      2\n",
       "21  ./test_set_2/0022.jpg      1\n",
       "22  ./test_set_2/0023.jpg      1\n",
       "23  ./test_set_2/0024.jpg      1\n",
       "24  ./test_set_2/0025.jpg      2\n",
       "25  ./test_set_2/0026.jpg      1\n",
       "26  ./test_set_2/0027.jpg      1\n",
       "27  ./test_set_2/0028.jpg      2\n",
       "28  ./test_set_2/0029.jpg      1\n",
       "29  ./test_set_2/0030.jpg      1\n",
       "30  ./test_set_2/0031.jpg      1\n",
       "31  ./test_set_2/0032.jpg      1\n",
       "32  ./test_set_2/0033.jpg      2\n",
       "33  ./test_set_2/0034.jpg      1\n",
       "34  ./test_set_2/0035.jpg      1\n",
       "35  ./test_set_2/0036.jpg      0\n",
       "36  ./test_set_2/0037.jpg      2\n",
       "37  ./test_set_2/0038.jpg      1\n",
       "38  ./test_set_2/0039.jpg      1\n",
       "39  ./test_set_2/0040.jpg      2\n",
       "40  ./test_set_2/0041.jpg      1\n",
       "41  ./test_set_2/0042.jpg      2\n",
       "42  ./test_set_2/0043.jpg      1\n",
       "43  ./test_set_2/0044.jpg      1\n",
       "44  ./test_set_2/0045.jpg      2\n",
       "45  ./test_set_2/0046.jpg      1\n",
       "46  ./test_set_2/0047.jpg      1\n",
       "47  ./test_set_2/0048.jpg      2\n",
       "48  ./test_set_2/0049.jpg      1\n",
       "49  ./test_set_2/0050.jpg      1\n",
       "50  ./test_set_2/0051.jpg      1\n",
       "51  ./test_set_2/0052.jpg      1"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[0:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'./test_set_2/0001.jpg'"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['case'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0001.jpg'"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os\n",
    "os.path.basename(df['case'][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>case</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0001.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0002.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0003.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0004.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0005.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0006.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0007.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0008.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0009.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0010.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0011.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0012.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0013.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0014.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0015.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0016.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0017.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0018.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0019.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0020.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0021.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0022.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0023.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0024.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0025.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0026.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0027.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>0028.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>0029.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0030.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>0031.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>0032.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>0033.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>0034.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>0035.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>0036.jpg</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>0037.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>0038.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>0039.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>0040.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>0041.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>0042.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>0043.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>0044.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>0045.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>0046.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>0047.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>0048.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>0049.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>0050.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>0051.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>0052.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        case  class\n",
       "0   0001.jpg      1\n",
       "1   0002.jpg      2\n",
       "2   0003.jpg      2\n",
       "3   0004.jpg      1\n",
       "4   0005.jpg      1\n",
       "5   0006.jpg      1\n",
       "6   0007.jpg      2\n",
       "7   0008.jpg      0\n",
       "8   0009.jpg      0\n",
       "9   0010.jpg      1\n",
       "10  0011.jpg      1\n",
       "11  0012.jpg      1\n",
       "12  0013.jpg      1\n",
       "13  0014.jpg      1\n",
       "14  0015.jpg      1\n",
       "15  0016.jpg      1\n",
       "16  0017.jpg      1\n",
       "17  0018.jpg      1\n",
       "18  0019.jpg      1\n",
       "19  0020.jpg      0\n",
       "20  0021.jpg      2\n",
       "21  0022.jpg      1\n",
       "22  0023.jpg      1\n",
       "23  0024.jpg      1\n",
       "24  0025.jpg      2\n",
       "25  0026.jpg      1\n",
       "26  0027.jpg      1\n",
       "27  0028.jpg      2\n",
       "28  0029.jpg      1\n",
       "29  0030.jpg      1\n",
       "30  0031.jpg      1\n",
       "31  0032.jpg      1\n",
       "32  0033.jpg      2\n",
       "33  0034.jpg      1\n",
       "34  0035.jpg      1\n",
       "35  0036.jpg      0\n",
       "36  0037.jpg      2\n",
       "37  0038.jpg      1\n",
       "38  0039.jpg      1\n",
       "39  0040.jpg      2\n",
       "40  0041.jpg      1\n",
       "41  0042.jpg      2\n",
       "42  0043.jpg      1\n",
       "43  0044.jpg      1\n",
       "44  0045.jpg      2\n",
       "45  0046.jpg      1\n",
       "46  0047.jpg      1\n",
       "47  0048.jpg      2\n",
       "48  0049.jpg      1\n",
       "49  0050.jpg      1\n",
       "50  0051.jpg      1\n",
       "51  0052.jpg      1"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['case']=df['case'].map(lambda a:os.path.basename(a))\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_file=fn+dt+\".csv\"\n",
    "df.to_csv(final_file,index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "        case  class\n",
      "0   0001.jpg      1\n",
      "1   0002.jpg      2\n",
      "2   0003.jpg      2\n",
      "3   0004.jpg      1\n",
      "4   0005.jpg      1\n",
      "5   0006.jpg      1\n",
      "6   0007.jpg      2\n",
      "7   0008.jpg      0\n",
      "8   0009.jpg      0\n",
      "9   0010.jpg      1\n",
      "10  0011.jpg      1\n",
      "11  0012.jpg      1\n",
      "12  0013.jpg      1\n",
      "13  0014.jpg      1\n",
      "14  0015.jpg      1\n",
      "15  0016.jpg      1\n",
      "16  0017.jpg      1\n",
      "17  0018.jpg      1\n",
      "18  0019.jpg      1\n",
      "19  0020.jpg      0\n",
      "20  0021.jpg      2\n",
      "21  0022.jpg      1\n",
      "22  0023.jpg      1\n",
      "23  0024.jpg      1\n",
      "24  0025.jpg      2\n",
      "25  0026.jpg      1\n",
      "26  0027.jpg      1\n",
      "27  0028.jpg      2\n",
      "28  0029.jpg      1\n",
      "29  0030.jpg      1\n",
      "30  0031.jpg      1\n",
      "31  0032.jpg      1\n",
      "32  0033.jpg      2\n",
      "33  0034.jpg      1\n",
      "34  0035.jpg      1\n",
      "35  0036.jpg      0\n",
      "36  0037.jpg      2\n",
      "37  0038.jpg      1\n",
      "38  0039.jpg      1\n",
      "39  0040.jpg      2\n",
      "40  0041.jpg      1\n",
      "41  0042.jpg      2\n",
      "42  0043.jpg      1\n",
      "43  0044.jpg      1\n",
      "44  0045.jpg      2\n",
      "45  0046.jpg      1\n",
      "46  0047.jpg      1\n",
      "47  0048.jpg      2\n",
      "48  0049.jpg      1\n",
      "49  0050.jpg      1\n",
      "50  0051.jpg      1\n",
      "51  0052.jpg      1\n"
     ]
    }
   ],
   "source": [
    "#import pandas\n",
    "df2 = pandas.read_csv(final_file)\n",
    "#df = pandas.read_csv(filename)\n",
    "print(df2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "\n",
      "       case\n",
      "class      \n",
      "0         4\n",
      "1        36\n",
      "2        12\n",
      "\n",
      "   class\n",
      "1     36\n",
      "2     12\n",
      "0      4\n"
     ]
    }
   ],
   "source": [
    "# label_encoder.inverse_transform(predClasses)\n",
    "print(type(df2))\n",
    "print()\n",
    "print(pandas.DataFrame(df2.groupby('class').count()))\n",
    "print()\n",
    "print(pandas.DataFrame(df2['class'].value_counts()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1   ./validation_2/normal\n",
      "2   ./validation_2/pneumonia\n",
      "0   ./validation_2/covid\n"
     ]
    }
   ],
   "source": [
    "for i,j in zip(df2['class'].unique(),label_encoder.inverse_transform(df2['class'].unique())):\n",
    "    print(i,' ',j)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1   36   ./validation_2/normal\n",
      "2   12   ./validation_2/pneumonia\n",
      "0   4   ./validation_2/covid\n"
     ]
    }
   ],
   "source": [
    "for i,j,k in zip(df['class'].unique(),df['class'].value_counts(),label_encoder.inverse_transform(df['class'].unique())):\n",
    "    print(i,' ',j,' ',k)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(class  class\n",
       " 0      0         4\n",
       " 1      1        36\n",
       " 2      2        12\n",
       " Name: class, dtype: int64,\n",
       " array([1, 2, 0], dtype=int64),\n",
       " array(['./validation_2/normal', './validation_2/pneumonia',\n",
       "        './validation_2/covid'], dtype='<U24'))"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('class')['class'].value_counts(),df['class'].unique(),label_encoder.inverse_transform(df['class'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}